Q3 2025 Tesla Inc Earnings Call

October 22, 2025

Yeah.

Okay.

Speaker #1: Because of the big difference between day and night usage . If you buffer that buffer , the energy with batteries , you can effectively double the energy output of the United States just with batteries pulling no incremental power plants .

Okay.

Thanks.

Okay.

Perfect.

Speaker #1: And . It's a . Difficult project , and it's worth noting that it's not like it's just automatic . I'm unaware of any robot program by Ford or GM or any of our US sort of car companies .

Okay.

Thank you.

Okay.

Provision.

Yes.

Okay.

Got it.

Yes.

Okay.

Speaker #1: And it's very difficult to build power plants. So, they take a long time. There's a lot of permitting, and it's not an industry that's used to moving fast.

[music].

Speaker #2: in real world No one can do real AI world AI can do with real . world AI . No one can do No one can do what we what can do we can with do real world with real AI world good insight .

Speaker #1: So we see the potential there for Tesla battery packs to greatly improve the energy output per year for any given grid, time.

Speaker #2: I general . think I think of that any Tesla has AI the highest out there in the intelligence car density of . Density of that is any only going to AI out there in get AI out there the in the car .

Speaker #2: full And fundamentally changing self-driving . And the nature of robotaxi . I . And think people just fundamentally changing the fundamentally nature of quite .

Okay.

[music].

Speaker #2: . So It's a , it's a . It's it's it's it's a a . Yeah . Cause cause the cars are all out there .

Speaker #2: . They're You know , because , you know , because the cause cars are because the all out cars there , are out there , they're they're , you know , we have millions that , you know , we have of cars millions of out cars out there their that that with a the software software update update become becomes self-driving cars .

Speaker #2: And in fact , with what with the advent with the advent of of clarity with on what we with with see now what we see as , now , as a achieving full self clarity driving , on unsupervised full self driving , I achieving should say full achieving full self self-driving , driving , unsupervised , full unsupervised , self self-driving , I should say confident in expanding I feel Tesla's .

Speaker #2: . Our future So production I was I was reticent to do that . . We had clarity . So I was I I was was I reticent to do was that reticent to do that .

Speaker #2: until we had Achieving clarity parity on on unsupervised full self-driving , but achieving achieving at this point unsupervised , full unsupervised self-driving . But at self-driving clarity this point , .

[music].

Speaker #1: Good afternoon , everyone , and welcome to Tesla's third Quarter 2025 Q&A webcast . My name is Travis Axelrod , head of Investor Relations , and I'm joined today by Elon Musk .

Speaker #1: Vaibhav Taneja and a number of other executives. Our Q3 results were announced at about 3 p.m. Central Time in the update deck we published at the same link as this webcast.

Speaker #1: During this call , we will discuss our business outlook and make forward looking statements . These comments are based on our predictions and expectations as of today .

Speaker #1: Actual events or results could differ materially due to a number of risks and uncertainties , and uncertainties , including those mentioned in our most recent filings with the SEC .

[music].

Speaker #1: We urge shareholders to read our definitive proxy statement , which contains important information about the matters to be voted on at the 2025 Annual Meeting .

Speaker #1: During the question and answer portion of today's call , please limit yourself to one question and one follow up . Please use the Raise Hand button to join the question queue .

Speaker #1: Before we jump into Q&A , Elon has some opening remarks . Elon . Elon Q&A has some Elon has opening some opening remarks .

Speaker #1: Remarks. Elon Musk. Okay...

Speaker #2: Thank you . We're I we're at thank you . a We're point for Tesla we're and at a we're at Ford critical as we inflection point bring for AI Tesla into the and our real strategy world and our going strategy forward as going we forward as bring AI we bring into the real into the real world .

Speaker #2: world . Tesla I think really is the it's leader important to in emphasize that AI Tesla really . Tesla is the really is leader the leader .

Speaker #2: into AI in I've general . I pretty good think pretty insight good insight into into Tesla has the I highest in AI in general .

Speaker #2: car And that is only , and going to get we're really better just the beginning of scaling at a at a . And we're really just at and really the beginning just the of beginning of scaling scaling , at driving quite robotaxi quite massively massively .

Speaker #2: The nature of transport . transport , I think which I people think just people just don't don't quite don't quite appreciate the degree to appreciate , take away which this which will this will take wear off .

Speaker #2: off . It's honestly , it's gonna be like a shock wave Where the where it's the honestly , it's it's gonna be like honestly it's a shock gonna be like a wave shock wave .

Speaker #2: And . Full full self-driving self cars driving cars . You know we're making and and and a couple million you know , we're a year you know we're making making a couple a million a couple year million a year with the advent of .

Speaker #2: I feel confident confident in in . So that expanding is Tesla's that is our production production . So that is expand . that is That is intent that is early as to we can expand as expand quickly as as we quickly as we can can .

Speaker #2: At this point , it I makes feel like sense . we've I feel got like we've got clarity and to it makes sense expand production .

Speaker #2: It makes sense to, as we reasonably can, expand production as fast as we were also able to, making huge strides. We're.

Speaker #2: We're also also making huge making huge , making huge impact on the . energy sector Make with making making impact on the impact energy on energy sector sector , with with so with with both battery Powerwall storage .

Speaker #2: storage and So especially with the with Megapack , both both Powerwall Powerwall and especially and with the especially

And welcome to Tesla's third quarter 2025, Q&A webcast my.

And welcome to Tesla's third quarter 2025, Q&A webcast.

My name is Travis Axelrod head of Investor Relations and I'm joined today by Elon Musk thereof to nurture and a number of other executives.

My name is Travis Axelrod head of Investor Relations and I'm joined today by Elon Musk live up to nurture and a number of other executives.

Our Q3 results were announced at about three P. M Central time in the update deck, we published at the same link as this webcast.

During this call we will discuss our business outlook and make forward looking statements. These comments are based on our predictions and expectations as of today.

Events or results could differ materially due to a number of risks and uncertainties.

[noise] uncertainties, including those mentioned in our most recent filings with the SEC.

Elon Musk: Time. It's a difficult project. It's worth noting that it's not just automatic. I'm unaware of any robot program by Ford or GM or any of our U.S. car companies. People, I think, maybe think of Tesla as a car company. We mostly make cars and battery packs. It's not just an obvious fall-off-a-log thing to make Optimus. We do have the ingredients of real-world AI and exceptional electrical and mechanical engineering capabilities and the ability to scale production, which I don't think anyone else has all of those ingredients. With version 14 of Full Self-Driving, you can see the reactions of people online. They're quite amazed. Actually, anyone in the U.S.

[noise] uncertainties, including those mentioned in our most recent filings with the SEC.

We urge shareholders to read our definitive proxy statement, which contains important information about the matters to be voted on at the 2025 annual meeting.

During the question and answer portion of today's call. Please limit yourself to one question and one follow up please use the raise hand button to join the question queue.

Before we jump into Q&A Elon has some opening remarks Elon.

Thank you Ware.

Okay.

We're at a critical inflection point for Tesla and our strategy going forward as we bring AI into the real world.

Okay.

We're at a critical inflection point for Tesla and our strategy going forward as we bring AI into the real world.

So, it's not like, it's not just like an obvious full of a log thing to make Optimus, but.

I think it's important to emphasize that because it really is the leader in real World AI.

But we do have the ingredients um, with of real world, AI.

I think it's important to emphasize that because it really is the leader in real World AI.

um,

No one can do what we can do with real world AI.

And exceptional electromechanical engineering capabilities.

No one can do what we can do with real world AI.

Um, and the ability to scale production.

I have pretty good insight into high AI in general I think that Tesla has the highest intelligence Betsy of any AI out there in the car.

Which I don't think anyone else has all of those ingredients.

Um,

so,

Uh, yeah. What

And that is only going to get better.

And we're really just at the beginning of scaling Adam quite massively full self driving and robo taxi and fundamentally changing the nature of transport having people just don't quite appreciate the degree to which this will.

with, with version 14 of

And we're really just at the beginning of scaling.

Adam quite massively full self driving and robo taxi.

And fundamentally changing the nature transport, having people just don't quite appreciate the degree to which this woe.

Elon Musk: can get version 14 if they just go and select, "I want the advanced software in their car." If you're listening right now and you'd like to try it out, just go in settings and say, "I want the advanced software," and you will get version 14. On the Megapack front, we unveiled Megablock, Megapack 3. We also have exciting plans for Megapack 4. Megapack 4 will incorporate a lot of what is normally in a substation and be able to output at probably 35 kilovolts directly. This greatly improves our ability to deploy Megapack because it's not dependent on building a substation up through 35 kV for Megapack 4. That'll be next. That's the engineering priority for Megapack. We look forward to unveiling Optimus V3, probably in Q1. I think it'll be ready to show off. That, I think, is going to be quite remarkable.

Take off.

Where the it's honestly, it's going be like a shockwave.

Where it's honestly, it's can be like a shockwave.

So.

the, uh, of of self-driving, which people you can see the reactions of, of people online. Um, but quite amazed, um, and actually anyone in the US can get a version 14, if they just go and select, uh, I want the advanced software in their car, so if if you're listening right now and you'd like to try it out, just uh, go and

Yeah.

It's.

Yeah.

Yeah, because it because the cars are all out there.

Um, in settings, if you say, "I want the advanced software," you will get version 14.

Yeah, because it because the cars are all out there.

Um,

You know there are millions of cars out there.

and uh,

You know there are millions of cars out there.

That's where the software update become.

yeah. So

That's where the software update become.

Full self driving cars.

And.

And you know we're making.

A couple of million a year.

And in fact with the advent of with what we see now is as a clarity on.

And in fact with the advent of with with what we see now is as a clarity on.

Uh, on the make pack front, we we we unveiled mega block Mega Pack 3. Um, we also have exciting plans for Mega Pack 4. Mega pack 4 will incorporate, uh, a lot of the um, uh, a lot of what is normally in a substation.

Achieving full self driving unsupervised full self driving I should say.

Uh, and be able to output, uh, at, uh,

I feel confident in expanding Tesla's production.

Probably 35 kilovolts. Uh, directly, so this...

So that is that is our intent to.

So that is that is our intent to expand.

Expand as quickly as we can our future production.

So I was curious since do that until we have clarity on.

Greatly improves our ability to deploy mega pack because it's not dependent on building, uh, a substation of 335 KV for Mega Pack 4, so that that'll be next. But that's the, that's the engineering priority for

Achieving.

Unsupervised full self driving but at this point I feel like we've got clarity and it makes sense too.

Um, mega pack. Um,

Unsupervised full self driving but at this point I feel like we've got clarity and it makes sense to us.

and uh,

we look forward to unveiling, Optimus V3.

Expand production as fast as we reasonably can.

Um, you know, probably in Q1, I think it'll be ready for.

We're also making huge.

To to show off.

and uh,

Make vacation impact on the energy sector with.

That's something I think is going to be quite remarkable. Um,

Elon Musk: It won't even seem like a robot. It'll seem like a person in a robot suit, which is kind of how we started off with Optimus. It'll seem so real that you'll need to poke it, I think, to believe that it's actually a robot. Obviously, the real-world intelligence we've developed for the car, most of that transfers to Optimus. It's a very good starting point. In conclusion, we're excited about the updated mission of Tesla, which is sustainable abundance. Going beyond sustainable energy to say sustainable abundance is the mission, where we believe with Optimus and Full Self-Driving that you can actually create a world where there is no poverty, where everyone has access to the finest medical care. Optimus will be an incredible surgeon, for example. Imagine if everyone had access to an incredible surgeon. Of course, we need to make sure Optimus is safe and everything.

With battery storage.

So with both the power wall, and especially with the Mega pack.

But the power wall, and especially with the Mega pack.

We are dramatically improving the.

If you, it won't even seem like a robot; it'll seem like a person in a robot suit, which is kind of how we started off with Optimus.

We are dramatically improving the ability to.

The ability to.

Um but it it'll seem so real.

Generate more energy from the grid.

that you'll need to like,

Generate more energy from the grid.

Let me sort of talk a little bit about that which is.

Poke it, I think to believe that it's actually a robot.

If you look at total U S energy.

If you look at total U S LNG.

um,

Okay. Buddy for example, there's roughly a terawatt of Gov.

Capability for example, there's roughly a terawatt of.

Continuous power available in the U S. But the average usage over a 24 hour cycle is only half of terawatt because of that.

And obviously, like the real world, it wasn't until we've developed for the car.

We have continuous power available in the U S. But the average usage over a 24 hour cycle is only half of terawatt, because the big difference between they're not usage.

Most of that transfers to optimism. So it's a very good starting point.

Different screen, they're not usage.

Um, in conclusion, uh,

If you bought for the buffer the energy with batteries you can effectively double the energy output in United States, just with batteries building no incremental power plants.

we're excited about the, you know, updated mission of Tesla, which is sustainable abundance.

um,

So, going beyond sustainable energy to say,

And it's very difficult to build pipelines.

Sustainable. Abundance is the mission where, uh,

And it's very difficult to build pipelines. So that takes a long time, there's a lot of permitting.

So they take a long time, there's a lot of permitting that's not.

we, we believe with

uh,

An industry that's used to moving fast so.

At an industry that's used to moving fast.

with Optimus and self-driving.

We see the potential therefore Tesla battery packs.

So.

Um,

We see the potential therefore, Tesla battery packs are too.

Two.

That, uh, you can actually create a world where there is.

Lately improve the.

No poverty.

Greatly improve the.

The energy output per year for any given grid U S or otherwise.

Uh, where everyone has access to the finest medical care. Um,

Okay.

We're also on the cusp of a.

Optimist will be an incredible surgeon.

We're also on the cusp of.

For example.

If something really tremendous with Optimus.

Which I think.

Um, and imagine if everyone had access to an incredible surgeon.

Which I think.

Um,

It's likely to be or has potential to be the biggest product of all time.

Is likely to be or has potential to be the biggest product of all time.

so, so, I think there's

Elon Musk: I do think we're headed for a world of sustainable abundance. I'm excited to work with the Tesla team to make that happen.

And.

You know, of course, we need to make sure Optimus is safe and everything, but...

Okay.

It's difficult project.

That's a difficult project.

But I, I do think we're headed for a world of sustainable abundance.

It's worth noting that it's not like it's just automatic.

And it's worth noting that it's not like it's just automatic.

And I'm excited to work with the Tesla team to make that happen.

I'm unaware of any robot program by a Ford.

Travis Axelrod: Great. Thank you very much, Elon. Vaibhav also has some opening remarks.

I'm unaware of any robot program by a Ford.

Ford or GM or by U S sort of car companies.

Ford or GM or sort of car companies.

Vaibhav Taneja: Thanks, Travis. Q3 was a special quarter at multiple levels. We set new records not just for deliveries and deployments, but also around a range of financial metrics from total revenues, energy gross profit, energy margins, to fresh free cash flow. This was the result of continued confidence of our customers in our products and the relentless efforts by the Tesla team. The strength in deliveries was attributed to strong performance across all regions. Greater China and APAC were up sequentially 33% and 29% respectively. North America was up 28%, while EMEA was up 25%. The pace in deliveries was the function of continued excitement around the new Model Y.

Great, thank you very much, Elon. Uh, Zebop also has some opening remarks.

Maybe think of Tesla as a car company were mostly make cars.

Thanks Travis.

Maybe think of Tesla as a car company were mostly make cars.

And battery packs.

But.

So it's not like it's not just like an obvious falloff of lug thing to make Optimus, but.

But we do have the ingredients.

Q3 was a special quarter at multiple levels. We set new records, not just for deliveries and deployments, but also around a range of financial metrics, from total revenues and gross profit to energy margins and free cash flow.

With real World AI.

And exceptional electrical mechanical engineering capabilities.

Tesla team.

Antibodies scale production.

Which I don't think anyone else has all of those ingredients.

So.

Yes.

With version 14 of.

The strengthening deliveries was attributed to strong performance across all regions, greater China, and APAC were up, sequentially, 33, and 29% respectively. North America was up 28% while Mia was up 25%.

With version 14 of.

The Oh self driving which people you can see the reactions of people online.

The upsell.

Vaibhav Taneja: We had previously talked about 2025 being the year of the Y and have since delivered on that promise with the new Model Y released in Q1, followed by Model Y long-wheelbase and performance, and more recently, Standard Y in North America and EMEA. We're now operating a robotaxi in two markets, Austin and most Bay Area cities. We've already expanded our coverage area in Austin three times since the initial launch and are on pace to continue expanding further. Unlike our competitors, our robotaxi fleet blends in the markets we operate in since they don't have extra sensor sets or peripherals which make them stick out. This is an underappreciated aspect of our current vehicle offerings, which are all designed for autonomous driving. We feel that as people experience the supervised FSD at scale, the demand for our vehicles, like Elon said, would increase significantly.

Self driving which people that you can see the reactions of people online.

The pace and deliveries were a function of continued excitement around the new Model Y.

But I'm quite amazed.

They're quite amazed.

It actually than anyone in the U S can get.

Actually than anyone in the U S can get a version 14, if they just go and select.

In 2014.

Go and select.

The advanced software in their car, so if you're listening right now and you'd like to try it out just going.

I want the advent software in their car. So if you're if you're listening right now and you'd like to try it out just to go in.

We had previously talked about 2025 being the year of the Y and it has since delivered on that promise with the new Model Y released in Q1, followed by the Model Y Long Range, Performance, and more recently, the Standard Y in North America.

And settings, and say I want to be advanced software and you'll get version 14.

Yeah.

And yes so.

On the <unk> front.

We are now operating a robotaxi in two markets: Austin and most Bay Area cities. We've already expanded our coverage area in Austin three times since the initial launch and are on pace to continue expanding further.

On the <unk> front.

Unveiled Mega block <unk> three mm. We also have exciting plans for Omega pack for Omega pack four will incorporate a lot of the.

We unveiled Mega block <unk> three mm.

We also have exciting plans for Omega pack for make.

<unk> four will incorporate.

Unlike our competitors, our robot taxi fleet blends in with the markets we operate in since they don't have extra sensor sets or peripherals that make them stick out.

The.

Okay.

Oh.

A lot of what is normally in a substation.

And be able to output.

Hi.

At.

Probably 35 kilovolt.

Directly so this.

Greatly improves our ability to deploy mega pack, because it's not dependent on building.

Vaibhav Taneja: On the FSD adoption front, we've continued to see decent progress. However, note that total paid FSD customer base is still small, around 12% of our current fleet. We're working with regulators in places like China and EMEA to obtain approvals so that we can get FSD in those regions as well. Now, covering a little bit on the financial side, automotive revenues increased 29% sequentially in line with the growth in deliveries. While regulatory credits declined sequentially, we entered into new contracts and continued delivery on previously entered contracts. Our automotive margins, excluding credits, increased marginally from 15% to 15.4%, which was attributed to improvements in material cost and better fixed cost absorption due to higher volumes. The energy storage business continued to deliver with record deployments, gross profit, and margins.

This is an underappreciated aspect of our current vehicle offerings, which are all designed for autonomous driving. We feel that as people experience the supervised FSD at scale, the demand for our vehicles, like Elon said, will increase significantly.

Our substation of 335 kv Omega pipe for so that that'll be thanks.

Our substation of 30 to 35 kv Omega pipe for so that that'll be thanks.

On the FSD adoption front, we've continued to see decent progress.

That's the best the engineering priority for.

Mega pack.

However, note that total paid FSD customer base is still small around 12% of our current Fleet.

Mega pack.

Yeah.

And.

Okay.

And.

We look forward to unveiling octopus b three.

Probably in Q1, I think it'll be ready for.

We're moving we're working with Regulators in places, like China and Amia to obtain approvals so that we can get FSD in those regions as well.

Probably in Q1, I think it'll be ready for.

So to show off.

And.

I don't think it's going to be quite remarkable.

That I don't think it's going to be quite remarkable.

Now, covering a little bit on the financial side, automotive revenues increased 29% sequentially, in line with the growth in deliveries.

If you would even seem like a robot it will seem like a person in the robot suite, which is kind of how we started off with autonomous.

If you would even seem like a robot it'll seem like a person in the robot suite, which is kind of how we started off with autonomous.

While regulatory credits declined sequentially, we entered into new contracts and continued delivery on previously entered contracts.

It will seem so real.

It'll seem so real.

That you'll need to like.

Okay, I think to believe that it's actually a robot.

Our automotive margins, excluding credits, increased marginally from 15% to 15.4%. This improvement was attributed to reductions in material costs and better fixed cost absorption due to higher volumes.

And obviously like the real World intelligence, where devote.

And obviously like they'd be the Royal Wolf intelligence, where devote.

Developed for the car.

We've developed for the car.

Vaibhav Taneja: As discussed before, this business has a bigger impact from tariffs, as measured by a percentage of COGS, since currently all sales procured are from China, while we're still working on other alternatives. However, as the ramp of Megapack Factory Shanghai is happening, this is helping us avoid tariffs because we are using this factory to supply the non-U.S. demand. Like Elon said, you know, grid-scale storage, the only way we can get to electricity fastest is by using storage. The other thing to keep in mind is we are seeing headwinds in this business given the increase in competition and tariffs. The total tariff impact for Q3 for both businesses was in excess of $400 million, generally split evenly between them. Services and other demonstrated a marked improvement sequentially. This was a function of improvements primarily in our insurance and service center businesses.

The energy storage business continued to deliver with record. Deployments gross profit and margins.

Most of that transfers to optimize or so.

Most of that transfers to optimists.

It's a very good starting point.

In conclusion.

As discussed before, this business has a bigger impact from terrorists, as measured by percentage of COGS.

In conclusion.

We're excited about the updated mission of Tesla, which is sustainable abundance.

Since currently all cells procured are from China, we are still working on other alternatives.

So going beyond sustainable energy to say sustain.

So going beyond sustainable energy to say sustainable.

Sustainable abundance as the mission where.

We believe with.

With Optimus and self driving.

However, as the ramp of Mega Factory Shanghai is happening, this is helping us avoid tariffs because we are using this factory to supply the non-U.S. demand.

But the optimists and self driving.

like Elon said, you know, grid scale storage

That's a you can actually create a world where there is.

The only way we can get to electricity faster is by using storage.

No poverty.

Where everyone has access to the finance medical care.

Where everyone has access to the finance.

The other thing to keep in mind is we are seeing headwinds in this business, given the increase in competition and tax.

Care.

Okay.

Great Optimists will be an incredible surgeon.

For example.

I'd imagine that everyone had access to an incredible surgeon.

And imagine if everyone had access to an incredible surgeon.

The total tariff impacts for Q3 2025 for both businesses were in excess of $400 million, generally split evenly between them.

So so I think there is.

So I think there is.

Now of course, we would make sure optimists as safe and everything but.

Now of course, we will make sure optimists as safe and everything but.

Vaibhav Taneja: Note that while small, our robotaxi costs are included within Services and other, along with our other businesses like paid Supercharging, Used Cars, parts and merchandise sales, etc. Our operating expenses increased sequentially. The largest increase included in restructuring and other related to certain actions undertaken to reduce costs and improve efficiency through convergence of our AI chip design efforts. Additionally, we incurred legal expenses related to proceedings in certain legal cases, as well as incremental costs incurred in preparation for our shareholder meeting. Such costs are recorded within SG&A. Further, our employee-related spend is increasing, especially in R&D, as we have recently granted various performance-based equity awards to employees working on AI initiatives, and therefore, such spend will continue to increase going forward.

But I do think we're headed for a world of sustainable abundance.

Services and other demonstrated a marked improvement sequentially. This was a function of improvements primarily in our insurance and Service Center. Businesses,

But I do think we're headed for a world of sustainable abundance.

And I'm excited to work with Tesla team to make that happen.

Great. Thank you very much Elon.

Great. Thank you very much Elon that up also has some opening remarks.

Note that while small, our robotaxi costs are included within services and other, along with our other businesses like paid supercharging, used car parts, and merchandise sales, etc.

We also have some opening remarks.

Thanks Thomas.

Our operating expenses, increased sequentially.

Q2 was a special quarter at multiple levels, we set new records not just for deliveries in deployments, but also around a range of financial metrics from total revenues and gross profit margins to free cash flow.

Q2 was a special quarter at multiple levels, we set new records not just for deliveries deployments, but also around a range of financial metrics from total revenues.

<unk> gross profit margins to free cash flow.

The largest increase included in restructuring in other related to certain actions undertaken to reduce costs and improve efficiency to convergence of our are AI. Chip design efforts.

This was the result of continued confidence of our customers and our products and the relentless efforts by the Tesla team.

Additionally, we incurred legal expenses related to proceedings in certain legal cases, as well as incremental costs.

The strength in deliveries was attributed to strong performance across all regions into China, and APAC were up sequentially 33, and 29% respectively. North America was up 28%, while EMEA was up 25%.

The strength in deliveries was attributed to strong performance across all regions of China, and APAC were up sequentially 33, and 29% respectively. North America was up 28%, while EMEA was up 25%.

Incurred interpretation for our shareholder meeting.

Such costs are recorded within SG&A.

The base in deliveries was the function of continued excitement around the new model y.

Based on deliveries was the function of continued excitement around the new model y.

Vaibhav Taneja: Our other income decreased sequentially, primarily from mark-to-market adjustments on BTC holdings, which was a much smaller gain of $80 million in Q3 versus $284 million in Q2, with the rest of the movement attributable to FX movements in the quarter. Our free cash flow for the quarter was approximately $4 billion, which was yet another record. Our total cash and investments at the end of the quarter were over $41 billion. On the CapEx front, while we are expecting to be around $9 billion for the current year, we're projecting the numbers to increase substantially in 2026 as we prepare the company for the next phase of growth in terms of not just our existing businesses, but our bets around AI initiatives, including Optimus. In conclusion, note that bringing AI into the real world is hard, but we have never shied away from doing what is hard.

Further, our employee related spend is increasing, especially in R&D as we have recently. Granted various performance base Equity Awards to employees, working on AI initiatives, and therefore, such spend will continue to increase going forward.

We have previously talked about 20 to 25 being the year of the why and have since delivered on that promise with the new model wise at least in Q1, followed my model Y long wheel base and performance and more recently standard why in North America and EMEA.

Previously you talked about 20 to 25 being the year of the why and have since delivered on that promise with the new model wise at least in Q1, followed by model Y long wheel base and performance and more recently standard why in North America and EMEA.

We are now operating a robo taxi in two markets Austin and most bay area cities.

On other income, our other income decreased sequentially, primarily from mark-to-market adjustments on BTC holdings, which reflected a much smaller gain of $80 million in Q3 versus $284 million in Q2. The rest of the movement was attributable to FX movements in the quarter.

We know operating a robo taxi in two markets Austin and most bay area cities.

We've already expanded our coverage area in Austin three times since the initial launch and are on pace to continue expanding further.

We're already expanded our coverage area in Austin three times since the initial launch and are on pace to continue expanding further.

Travis Axelrod: Good afternoon, everyone, and welcome to Tesla's Q3 2025 Q&A webcast. My name is Travis Axelrod, Head of Investor Relations, and I'm joined today by Elon Musk, Vaibhav Taneja, and a number of other executives. Our Q3 results were announced at about 3:00PM Central Time in the update deck we published at the same link as this webcast. During this call, we will discuss our business outlook and make forward-looking statements. These comments are based on our predictions and expectations as of today. Actual events or results could differ materially due to a number of risks and uncertainties, including those mentioned in our most recent filings with the SEC. We urge shareholders to read our definitive proxy statement, which contains important information about the matters we voted on at the 2025 annual meeting.

Our total cash and Investments at the end of the quarter were over 41 billion.

Unlike our competitors.

Like our competitors undergo taxi fleet blends in the markets, we operate and since they don't have extra census at Outback.

See fleet blends in the markets, we operate and since they don't have extra sensors that's.

I'll play defense, which make them stick out.

Paddy fields, which make them stick out.

This is an underappreciated aspect of our current wake offerings, which are all designed for autonomous driving.

This is an underappreciated aspect of our current wait got offerings, which are all designed for autonomous driving.

On the capex front, while we are expecting to be around $9 billion for the current year, we're projecting the numbers to increase substantially in 2026 as we prepare the company for the next phase of growth. This includes not just our existing businesses, but also our bets around AI initiatives, including Optimist.

Feel that his experience.

As people experience the supervisor FSD at scale the demand for vehicles like Elon said would increase significantly.

As people experience the supervised FSD at scale the demand for vehicles like Elon said would increase significantly.

In conclusion, note that bringing AI into the real world is hard.

Vaibhav Taneja: We are extremely excited about the future and are laying down the foundation, the benefits of which will be realized over years to come. I would like to end by thanking the Tesla team, our customers, our investors, and supporters for the continued belief in us.

But we have never shied away from doing what is hard.

On the SSD adoption front.

Continue to see decent progress.

We are extremely excited about the future and are laying down the foundation. The benefits of which will be realized over years to come.

Continue to see decent progress.

I would note that total paid SSD customer base is still small around 12% of our current fleet.

Travis Axelrod: During the question-and-answer portion of today's call, please limit yourself to one question and one follow-up. Please use the raise hand button to join the question queue. Before we jump into Q&A, Elon has some opening remarks. Elon?

We're moving we're working with regulators in places like China, and then we have to obtain approvals. So that we can get efficacy in those regions as well.

I would like to end by thanking the Tesla team, our customers, our investors, and supporters for the continued belief in us.

Travis Axelrod: Thank you very much, Vaibhav. Now let's go to investor questions. From say.com, the first question is, what are the latest robotaxi metrics, fleet size, cumulative miles, rides completed, intervention rates, and when will safety drivers be removed? What are the obstacles still preventing unsupervised FSD from being deployed to customer vehicles?

We're moving we're working with regulators in places like China, and EMEA to obtain approvals. So that we can get efficacy in those regions as well.

Now covering a little bit on the financial side.

Automotive revenues increased 29% sequentially in line with the growth in deliveries.

Total revenues increased 29% sequentially in line with the growth in deliveries.

Elon Musk: Thank you. We're at a critical inflection point for Tesla and our strategy going forward as we bring AI into the real world. I think it's important to emphasize that Tesla really is the leader in real-world AI. No one can do what we can do with real-world AI. I have pretty good insight into AI in general. I think that Tesla has the highest intelligence density of any AI out there in the car, and that is only going to get better. And we're really just at the beginning of scaling at a quite massively full self-driving and robotaxi and fundamentally changing the nature of transport. I think people just don't quite appreciate the degree to which this will take off. Honestly, it's going to be like a shockwave.

Thank you very much, bye-bye. Now let's go to investor questions from say.com. The first question is, what are the latest Robo taxi? Metrics? Fleet size cumulative, miles. Rides completed intervention rates, and when will Safety Drivers be removed.

While regulatory cause its declined sequentially.

What are the obstacles? Still preventing unsupervised FSD from being deployed to customer vehicles?

Elon Musk: I'll start off with that, and then Ashok can elaborate. We are expecting to have no safety drivers in at least large parts of Austin by the end of this year. Within a few months, we expect to have no safety drivers at all, at least in parts of Austin. We're obviously being very cautious about the deployment. Our goal is to be actually paranoid about deployment because obviously even one accident will be front-page headline news worldwide. It's better for us to take a cautious approach here. We do expect to have no safety drivers in the car in Austin within a few months. I think that's perhaps the most important data point. We do expect to be operating robotaxi in, I think, about 8 to 10 metro areas by the end of the year. It depends on various regulatory approvals.

We entered into new contracts and continued delivery on previously entered contacts.

I'll, I'll start off with that and then shock and elaborate. Um,

Our automotive margins, excluding credits increased marginally from 15% to $15, four which was attributed to improvements in material cost and better fixed cost absorption due to higher volumes.

but uh, we we are expecting to

Our energy storage business continued to deliver with record deployment gross profit and margins.

Energy storage business continued to deliver with record deployment gross profit and margins.

As discussed before this business has a bigger impact from tariffs as measured by a percentage of Cogs.

Have uh no Safety drivers in uh at least um large parts of Austin by the end of this year. So within a few months, uh, we expect to have no Safety Drivers um at all um and at least in parts of Boston we're obviously being very cautious about the deployment. So um,

Since currently all self procured from China, while we're still working on other alternatives.

However, as.

So our goal is to be actually paranoid about deployment because obviously even 1 accident will be front page, headline news worldwide. So, um, you know, it's it's better for us to take a cautious approach here.

The ramp of my guess factory Shanghai is happening this is helping us a lot of tenants because we are using this factory to supply the non U S demand.

The ramp on some I guess factory Shanghai is happening this is helping us a lot of tenants because we are using this factory to supply the non U S demand.

Um but we do expect to have no no no safe drivers in the car. Um

Elon Musk: So because the cars are all out there, there are millions of cars out there that with a software update become Full Self-Driving cars. And we're making a couple million a year. And in fact, with the advent of what we see now as a clarity on achieving Full Self-Driving, unsupervised Full Self-Driving, I should say, I feel confident in expanding Tesla's production. So that is our intent to expand as quickly as we can our future production. So I was reticent to do that until we had clarity on achieving unsupervised Full Self-Driving. But at this point, I feel like we've got clarity, and it makes sense to expand production as fast as we reasonably can. We're also making a huge impact on the energy sector with battery storage.

Like Elon said, you know good scale storage.

The only way we can get to electricity fastest is by using storage.

In Austin, uh, within a few months, I think that's perhaps the most important, uh, data point. And then we do expect to be, uh, operating a Robo taxi in.

uh, I think about, uh,

The other thing to keep in mind is we are seeing headwinds in this business given the e&ps and competition and tests.

The other thing to keep in mind is we are seeing headwinds in this business given the MPS and competition and tests.

8 to 10 metro areas by the end of the year.

Elon Musk: You can actually, I think, most of our regulatory applications are online. You can kind of see them because they're public information. We expect to be operating in Nevada and Florida and Arizona by the end of the year. Ashok?

Um, you know, it depends on various regulatory approvals. Um,

The total tariff impacts for Q3 for both businesses was in excess of $400 million generally split evenly between them.

So this is another demonstrated a marked improvement sequentially. This was a function of improvements primarily in our insurance and service center businesses.

Services. Another demonstrated a marked improvement sequentially. This was a function of improvements primarily in our insurance and service center businesses.

And um, but you, you can actually I think most of our regulatory applications are online. You can kind of see them uh because they're public information. Um but we expect to be operating in Nevada and Florida and Arizona. Um, by the end of the year,

Ashok Elluswamy: Yeah. We continue to operate our fleet in Austin without anyone in the driver's seat, and we have covered more than a quarter million miles with that. In the Bay Area, where we still have a person in the driver's seat because of the regulations, we cross more than a million miles. We continue to see that the robotaxi fleet works really well. Customers are really happy, and there's no notable issues. On the customer side, customers have used Full Self-Driving supervised for a total of 6 billion miles as of yesterday. That's a big milestone. Overall, the safety continues to be very good. As Elon mentioned, we are on track to remove the person from inside the car altogether, starting with Austin.

Um, for sure.

Note that while small are what actually costs are included within services. Another along with our other businesses like pay supercharging used car parts and merchandise sales et cetera.

Note that while small our robo taxi costs are included within services another along with our other businesses like page Supercharging used car parts and merchandize sales et cetera.

Our operating expenses increased sequentially.

Yeah, we we continue to operate our um, Fleet in Austin, without anyone in the driver's seat and we have covered more than a quarter million miles, um, with that. And then in the Bay Area, where we still have a person in the driver's seat because of the regulations, we uh, cross more than a million miles.

The largest increase included in restructuring and other related to certain actions undertaken to reduce cost and improve efficiency to convergence. So far yeah, yeah chip design efforts.

The largest increase included in restructuring and other related to certain actions undertaken to reduce cost and improve efficiency to convergence of our yeah, Yeah chip design efforts.

Um, so we continue to see that the fleet, um, robotics fleet works really well. Customers are really happy. Um,

Italy, being legal expenses related to proceedings, and certain legal cases, as well as incremental costs incurred.

Additionally, we incurred legal expenses related to proceedings, and certain legal cases, as well as incremental costs incurred.

Elon Musk: So with both Powerwall and especially with the Megapack, we are dramatically improving the ability to generate more energy from the grid. So let me sort of talk a little bit about that, which is if you look at total US energy capability, for example, there's roughly a terawatt of continuous power available in the US. But the average usage over a 24-hour cycle is only half a terawatt because of the big difference between day and night usage. If you buffer the energy with batteries, you can effectively double the energy output of the United States just with batteries building no incremental power plants. And it's very difficult to build power plants. So they take a long time. There's a lot of permitting, and it's not an industry that's used to moving fast.

In preparation for our shareholder meeting such costs are recorded within SG&A.

Further our implied related spend is increasing especially in R&D as Ian recently granted various performance based equity awards to employees working on AI initiatives and of course that spend will continue to increase going forward.

And understand notable issues on the customer. Uh, side we have customers have used FSD supervised, for a total of 6 billion miles, um, as of yesterday. Um, so that's like a big milestone, uh, and overall the safety continues to be very good. And as Elon mentioned, we are on, um, on track to remove the person from inside the car all together, uh, starting with Austin.

Travis Axelrod: Great. The next question is, what is the demand and backlog for Megapack, Powerwall, solar, or energy storage systems? With the current AI boom, is Tesla planning to supply power to other hyperscalers?

On the other income or other income decreased sequentially, primarily from mark to market adjustments on BTC holdings, which was much smaller gain of $80 million in Q3 versus $284 million in Q2, but the rest of the movement attributable to FX movements in the quarter.

On the other income or other income decreased sequentially, primarily from mark to market adjustments on BTC holdings, which was much smaller gain of $80 million in Q3 versus 284 million in Q2, but the rest of the movement attributable to FX movements in the quarter.

Great. Uh, the next question is: What is the demand and backlog for Mega Pack, Powerwall, solar, or energy storage systems with the current AI boom? Is Tesla planning to supply power to other hyperscalers?

Lars Moravy: Thanks. Demand for Megapack and Powerwall continues to be really strong into next year. We received very strong positive customer feedback on our Megapack product, which will begin shipping next year out of Houston. We're seeing remarkable growth in the demand for AI and data center applications as hyperscalers and utilities have seen the versatility of the Megapack product to increase reliability and relieve grid constraints, as Elon was talking about. We've also seen a surge in residential solar demand in the U.S. due to policy changes, which we expect to continue into the first half of 2026 as we introduce a new solar lease product. We also began production of our Tesla residential solar panel in our Buffalo factory, and we will be shipping that to customers starting Q1. The panel has industry-leading aesthetics and shade performance and demonstrates our continued commitment to U.S. manufacturing.

Our free cash flow for the quarter was approximately $4 billion, which was yet another record.

Thanks. Demand for the Mega Pack and Powerwall continues to be really strong into next year. We received very strong positive customer feedback on our Mega Block product, which will begin shipping next year out of Houston.

Our free cash flow for the quarter was approximately $4 billion, which was yet another record.

Our total cash and investments at the end of the quarter were over 41 billion.

And we're seeing remarkable growth in the in the demand for AI and data center applications, as hyperscalers and utilities have seen the versatility of the mega pack product.

On the Capex front, while we are expecting to be around 9 billion for the current year.

On the Capex front, while we are expecting to be around 9 billion for the current year. We are projecting the numbers to increase substantially in 2026 as we prepare the company for the next phase of growth in terms of not just the existing businesses without bets around air initiatives, including Optimus.

Elon Musk: So we see the potential there for Tesla battery packs to greatly improve the energy output per year for any given grid, US or otherwise. We're also on the cusp of something really tremendous with Optimus, which I think is likely to be or has potential to be the biggest product of all time. And it's a difficult project. And it's worth noting that it's not like it's just automatic. I'm unaware of any robot program by Ford or GM or any of our US sort of car companies. People like, I think, maybe think of Tesla as a car company. We mostly make cars and battery packs.

To increase reliability and relieve grid constraints, as Elon was talking about.

We're projecting the numbers to increase substantially in 2026 as we prepare the company for the next phase of growth in terms of not just the existing businesses, but not bets around initiatives, including Optimus.

We've also seen a surge in residential solar demand in the U.S. due to policy changes, which we expect to continue into the first half of 2026.

As we introduced a new solar lease product.

In conclusion, not that binging, yeah into real World is hot.

In conclusion note that bringing <unk> into real world is hot.

And we also began production of our Tesla residential solar panel in our Buffalo factory, and we will be shipping that to customers starting Q1.

But we have never shied away from doing what is hot.

We are extremely excited about the future.

We are extremely excited about the future and are laying down the foundation, the benefits of which will be realized or years to come.

The panel has industry-leading Aesthetics and shade performance, and demonstrates. Our continued commitment to us, Manufacturing,

Laying down the foundation, the benefits of which will be realized or years to come.

Travis Axelrod: Great. Thank you, Mike. Unfortunately, the next question is related to future products. This is not the appropriate venue to cover that. We're going to have to skip it. The question after that is, what are the present challenges in bringing Optimus to market, considering app control software, engineering hardware, training general mobility models, training task-specific models, training voice models, implementing manufacturing, and establishing supply chains?

Great. Thank you. Mike.

I would like to end by thanking the Tesla team our customers, our investors and support us for the continued belief in us.

Thank you very much my boss now, let's go to investor questions.

Thank you very much my boss now, let's go to Investor questions from say Dot Com. The first question is what are the latest robo taxi metrics fleet size cumulative miles rides completed intervention rates and when will safety drivers be removed.

<unk> Dot com. The first question is what are the latest robo taxi metrics fleet size cumulative miles rides completed intervention rates and when will safety driver he removed.

Elon Musk: But so it's not just like an obvious slam-dunk thing to make Optimus, but we do have the ingredients of real-world AI and exceptional electrical and mechanical engineering capabilities, and the ability to scale production, which I don't think anyone else has all of those ingredients. So yeah, with version 14 of self-driving, which, people, you can see the reactions of people online, they're quite amazed. Actually, anyone in the US can get version 14 if they just go and select, "I want the advanced software in their car." So if you're listening right now and you'd like to try it out, just go in settings and say, "I want the advanced software," and you will get version 14. And yeah, so on the Megapack front, we unveiled Megablock, Megapack 3. We also have exciting plans for Megapack 4.

This is related to future products. Uh, this is not the appropriate venue to cover that, so we're going to have to skip it. Um, the question after that is: what are the present challenges in bringing optimists to market, considering app control, software engineering, hardware training in general, mobility models, training task-specific models, training voice models, implementing manufacturing, and establishing supply chains?

What are the obstacles still preventing unsupervised FSD from being deployed to customer vehicles.

Elon Musk: Yeah. I mean, bringing Optimus to market is an incredibly difficult task, to be clear. It's not like some walk in the park at some point. I mean, actually, technically, Optimus can walk in the park right now. We do have Optimus robots that walk around our offices at our engineering headquarters in Palo Alto, California, basically 24 hours a day, seven days a week. Any visitors that come by, you actually can stop one of the Optimus robots and ask it to take you somewhere, and it'll literally take you to that meeting room or that location in the building. I don't want to downplay the difficulty of Optimus. It's an incredibly difficult thing. Especially, it's difficult to create a hand that is as dexterous and capable as the human hand, which is incredible. The human hand is an incredible thing.

I'll start off with that and then shot can elaborate.

I'll start off with that and then Chuck can elaborate.

But.

Yeah. I mean bringing office Optimist Market is an incredibly difficult task, to be clear. It's not like some walk in the park. Um,

But.

We are expecting to.

I have no safety drivers and at least.

At some point. I mean at this actually technically Optimus can walk in the park right now.

Have no safety drivers and at least.

Large parts of Austin by the end of this year. So within a few months, we expect to have no safety drivers.

At all.

Eastern parts, Boston, we're obviously being very cautious about.

At least in parts Boston, we're obviously being very cautious about the deployment so.

Um, we do have Optimus robots that walk around our offices at our engineering headquarters in Palo Alto, California, basically 24 hours a day, 7 days a week. Um, so any visitors that come by, you actually um,

The deployment so.

So our goal is to actually be paranoid about deployment, because obviously, even one accident will be front page headline news worldwide.

You can you can stop to stop 1 of The Optimist robots and uh ask you to take you somewhere and it'll literally take you to that meeting room or that location in the building.

um,

So.

so, um,

No.

Yes, that's better for us to take a cautious approach here.

I don't want to downplay the difficulty of Optimus. It's uh,

But we do expect to have no no safe drivers in the car.

But we do expect to have no no six drivers in the car.

And Austin.

It's an incredibly difficult thing, especially; it's difficult to create a um,

And Austin.

Within a few months.

I think that's perhaps the most important.

That's perhaps the most important.

At a point and then we do expect to be operating.

Elon Musk: The more you study the human hand, the more incredible you realize the human hand is and why you need five, you know, four fingers and the thumb, why the fingers have certain degrees of freedom, why the various muscles are of different strengths, and the fingers are of different lengths. It turns out, actually, that those are all there for a reason. Making the hand and forearm, because most of the actuator, just like the human hand, the muscles that control your hand are actually primarily in your forearm. The Optimus hand and forearm is an incredibly difficult engineering challenge. I'd say it's more difficult than the rest of, from an electromechanical standpoint, the forearm and hand is more difficult than the entire rest of the robot. In order to have a useful generalized robot, you do need this. You do need an incredible hand.

At a point and then we do expect to be operating.

Robo taxi.

In.

I think about.

I think about eight.

Eight to 10 metro areas by the end of the year.

It depends on.

A hand that is as dextrous and capable as the human hand, which is an incredible. The human hand is an incredible thing. Um, that the more you study, the human hand, the more incredible. You realize the human hand is and and why you need 5, you know, 4 fingers in the thumb. Why the why the fingers have certain degrees of freedom? Um,

It depends on various regulatory approvals.

There is regulatory approvals.

Elon Musk: Megapack 4 will incorporate a lot of what is normally in a substation and be able to output at probably 35kV directly. So this greatly improves our ability to deploy Megapack because it's not dependent on building a substation up through 35kV for Megapack 4. So that'll be next. That's the engineering priority for Megapack. And we look forward to unveiling Optimus V3 probably in Q1. I think it'll be ready to show off. And that, I think, is going to be quite remarkable. It won't even seem like a robot. It'll seem like a person in a robot suit, which is kind of how we started off with Optimus. But it'll seem so real that you'll need to poke it, I think, to believe that it's actually a robot. And obviously, the real-world intelligence we've developed for the car, most of that transfers to Optimus.

And but you can actually I think most of our regulatory applications or online you can kind of see them because they're public information.

And but you can actually I think most of our regulatory applications or online you can kind of see them because they are public information.

You know, why the various muscles are of different strengths, um, the fingers are of different lengths, um, and uh,

But we expect to be operating in Nevada, and Florida and Arizona.

it turns out actually that all those are all there for a reason. Um,

But we expect to be operating in Nevada, and Florida and Arizona.

and uh,

By the end of the year.

Sure.

Yes, we are continuing to operate our.

Yes.

Continuing to operate our.

<unk> fleet in Austin.

Fleet in Austin without anyone in the driver seat and be I'll call it more than a quarter million lines.

In the driver's seat.

I would call it more than a quarter million lines.

And then in the Bay area.

So, making making an that the hand and, and forearms, because most of the most of the actuators are just like the human hand, the the muscles of that control, your hand are actually primarily in your forearm. Um, The Optimist hand and forearm is an incredibly difficult engineering challenge. Um, it's I'd say it's

The person that I have received the recognition.

Didn't have a person that I haven't seen.

Missions.

More than a million miles.

Across more than a million miles.

So and we continue to see that the feet.

So and we continue to see that defeat.

a more difficult than the rest of from an electrical mechanical standpoint. The forearm and hand are, it's more difficult than the entire rest of the robot.

It looks really bad that's what they're really happy.

um,

It was really bad customers that really happy.

And there's no notable issues on the customer side yet.

And there's no notable issues on the customer side, yet customers have used <unk> for a total of 6 billion miles.

Elon Musk: You need the real-world AI. You need to be able to scale up that production to have it be relevant because it's not relevant if it's just a few hundred robots. You need to be able to make Optimus robots at volumes comparable to vehicles, if not significantly higher. If you're trying to make a million or something per year, trying to make a million Optimus robots per year, that manufacturing challenge is immense, considering that the supply chain doesn't exist. With cars, you've got an existing supply chain. With computers, you've got an existing supply chain. With a humanoid robot, there is no supply chain. In order to manufacture that, Tesla actually has to be very vertically integrated and manufacture very deep into the supply chain, manufacture the parts internally because there just is no supply chain.

<unk> provides for a total of 6 billion miles.

So, I've been really in order to have a useful generalized robot use. You do need this; you do need an incredible hand.

And, uh, and then you need the real-world AI.

Yesterday.

Um,

Yesterday.

So that's like a big milestone and all these empty continues to be very good and I think you mentioned.

So that's like a big milestone and all these safety continues to be very good.

and you need to go to scale up that production.

Are on track to do more of the person from inside the car altogether, starting with Boston.

Mentioned.

Hum.

On track to do more of the person from inside the car altogether, starting with Boston.

Um, to have it be relevant because it's not relevant if it's just a few hundred robots. Um, but you need to be able to make.

Optimus robots at volumes comparable to vehicles, um,

Great.

Yeah.

The next question is what is the demand and backlog for Mega pack power wall solar or energy storage systems with the current AI done.

Right.

if not significantly higher, um,

The next question is what is the demand and backlog for Mega pack power wall solar or energy storage systems with the current AI.

So, uh, if you're trying to make a million or something, per year, a tournament, Optimus robots per year.

Is tesla playing to supply power to other hyperscale or.

Is tesla planning to supply power to other hyperscale.

That manufacturing challenge is immense, considering that the supply chain doesn't exist.

Demand for makeup back in power wall continues to be really strong into next year. We received very strong positive customer feedback on our Mega Black Mega block product, which will begin shipping next year not a houston.

Elon Musk: So it's a very good starting point. In conclusion, we're excited about the updated mission of Tesla, which is sustainable abundance. So going beyond sustainable energy to say sustainable abundance is the mission, where we believe with Optimus and self-driving that you can actually create a world where there is no poverty, where everyone has access to the finest medical care. Optimus will be an incredible surgeon, for example. And imagine if everyone had access to an incredible surgeon. So I think there's now, of course, we need to make sure Optimus is safe and everything. But I do think we're headed for a world of sustainable abundance. And I'm excited to work with the Tesla team to make that happen.

Thanks demand for makeup pack and power wall continues to be really strong into next year. We received very strong positive customer feedback on our Mega Mega block product, which will begin shipping next year out of Houston.

So, with cars, you've got an existing supply chain. With computers, you've got an existing supply chain with, uh...

With a humanoid robot, there is no supply chain.

Um,

And we're seeing remarkable growth in the demand for AI and data center applications as Hyperscale and utilities have seen the versatility of the Mega pack product.

So, in order to manufacture, Tesla actually has to be,

And we're seeing remarkable growth in the demand for AI and data center applications as hyper scaler and utilities have seen the versatility of the Mega pack product.

Very robotically integrated, uh, and manufactured.

um,

To increase reliability and received and relieve grid constraints as you'd want us talking about.

To increase reliability and received and relieve grid constraints as he was talking about.

Very deep into the supply chain, we manufacture the parts internally because there just is no supply chain.

Elon Musk: This is the kind of thing where I'm like, if I put myself in the position of a startup trying to make a humanoid robot, I'm like, I don't know how to do it without an immense amount of manufacturing technology. That's why I think Tesla is in almost a unique, I think, unique position when you consider manufacturing technology, scaling, real-world AI, and a truly dexterous hand. Those are generally the things that are missing when you read about other robots, that they just don't have those three things. I think we can achieve all those things, those three things, with an immense amount of work. That is the game plan. My fundamental concern with regard to how much voting control I have in Tesla is, if I go ahead and build this enormous robot army, can I just be ousted at some point in the future?

Um,

We've also seen a surge in residential solar demand in the U S. Due to policy changes, which we expect to continue into the first half of 2026.

We've also seen a surge in residential solar demand in the U S. Due to policy changes, which we expect to continue into the first half of 2026, as we introduced a new solar lease product.

We introduced a new solar lease product.

And we also began production of our test of residential solar panel and our Buffalo factory, and we will be shipping that to customers starting in Q1 the.

So, uh, this is this is the kind of thing where I'm like if I put myself in the position of a startup trying to make a humanoid robot, I'm like, I don't know how to do it without an immense amount of manufacturing technology.

And we also began production of our Tesla residential solar panel and our Buffalo factory, and we will be shipping that to customers starting in Q1.

um,

so,

The panel has industry, leading aesthetics and shade performance and demonstrates our continued commitment to U S manufacturing.

That's that's why I think like Tesla's in some almost a unique. I think I think unique position when you consider manufacturing technology scaling um real world AI.

And a truly dextrous hand.

Great. Thank you Mike.

And Unfortunately, the next question is related to future products. This is not the appropriate venue to cover that so we're going to have to skip it.

And Unfortunately, the next question is it related to future products and this is not the appropriate venue to cover that so we're going to have to skip it.

Uh, those are generally the things that are missing when you read about, um, other robots that they just don't have those three things.

um,

The question after that is what does it present challenges in bringing optimists to market considering App control software engineering hardware training General mobility models training task specific models training voice models implementing manufacturing and establishing supply chains.

So now I think we can.

On the question after that is what does it present challenges in bringing optimists to market considering App control software engineering hardware training General mobility models training task specific models training voice models implementing manufacturing and establishing supply chains.

Travis Axelrod: Great. Thank you very much, Elon. Vaibhav also has some opening remarks.

Achieve all those things, those three things, with an immense amount of work.

Vaibhav Taneja: Thanks, Travis. Q3 was a special quarter at multiple levels. We set new records not just for deliveries and deployments, but also around a range of financial metrics from total revenues, energy gross profit, energy margins, to free cash flow. This was the result of continued confidence of our customers in our products and the relentless efforts by the Tesla team. The strength in deliveries was attributed to strong performance across all regions. Greater China and APAC were up sequentially 33% and 29%, respectively. North America was up 28%, while EMEA was up 25%. The pace in deliveries was the function of continued excitement around the new Model Y.

And, um, and that, that is the Game Plan.

so, um,

you know my my my

You know, fundamental concern with regard to

Yes, I mean, bringing office after Mr market is an incredibly difficult task to be clear, it's not like some walk in the park.

Yes, I mean, bringing up is of utmost market is an incredibly difficult task to be clear, it's not like some walk in the park.

How much voting control do I have in Tesla?

Elon Musk: That's my biggest concern. That is really the only thing I'm trying to address with this, so what's called compensation, but it's not like I'm going to go spend the money. It's just, if we build this robot army, do I have at least a strong influence over that robot army? Not control, but a strong influence. That's what it comes down to in a nutshell. I don't feel comfortable building that robot army if I don't have at least a strong influence.

If I go ahead and build this enormous robot army, can I just be outside at some point in the future?

At some point I mean at this actually technically octopus can walk in the park right now.

And we do have autonomous robots that.

Walk around our offices at our engineering headquarters in Palo Alto, California.

it's just

Basically 24 hours a day seven days a week.

You know, if we build this robot army, um,

Basically 24 hours a day seven days a week.

So anybody that can buy you actually.

So any visitors that come by actually.

You can stop to stop one of the Optimus robots and ask him to take you somewhere and it'll literally take you to that meeting room or their location in the building.

You can stop to stop one of the optimists robots and ask him to take you somewhere and it'll literally take you to that meeting room or their location in the building.

Do I have at least, uh, a strong influence over that robot? Not control, but a strong influence.

That's what it comes down to, in a nutshell. Um, like I don't feel comfortable building that robot army if I don't have at least a strong influence.

So.

<unk>.

So.

Travis Axelrod: Great. Thank you. We've already covered robotaxi expansion. Unfortunately, the question after that is another future product question. We're going to have to skip that. The next one is, can you update us on the $16.5 billion Samsung chip deal in Taylor? Given the importance of semiconductors to autonomy in Tesla's AI-driven future, what gives you confidence Samsung can fulfill AI-5 at Tesla's timelines and achieve relatively better yields and costs versus TSMC?

I want to downplay the difficulty partners side.

I don't want to downplay the difficulty partners at site.

It's an incredibly difficult thing, especially.

It's an incredibly difficult thing, especially it's difficult to create a.

Difficult to create a.

And that is as dexterous and capable as the human hand, which is an incredible human hand is an incredible thing.

And that is as dexterous and capable of the human hand, which is an incredible human how does an incredible thing.

The more you studied the human hands the more incredible you realize the human hand is and why you need five four fingers and a thumb why the why the fingers have certain degrees.

But the more you studied the human hands more incredible you realize the human hand is and.

And why you need five.

Four fingers and a thumb why the why the fingers have certain degrees of freedom.

Freedom.

You know what why the various muscles are different strengths the fingers are a different lengths.

You know what why the.

Elon Musk: Okay. I'm going to give quite a long answer to this question because I have to unpack this question and then answer the unpacked version. First of all, I have nothing but great things to say about Samsung. They're an amazing company. Samsung, it is worth noting, does manufacture our AI-4 computer and does a great job doing that. Now with the AI-5, and here's where I need to make a point of clarification relative to some comments I've made publicly before, we're actually going to focus both TSMC and Samsung initially on AI-5. The AI-5 chip design by Tesla is, I think it's an amazing design. I've spent almost every weekend for the last few months with the chip design team working on AI-5. I don't hand out praise easily, but I have to say that I think the Tesla chip team is really designing an incredible chip here.

Great, thank you. Uh, we've already covered Robo taxi expansion. Unfortunately, the question after that is another future product question, so we're going to have to skip that. Um, the next one though is, can you update us on the 1, 6, 5, 7?

Muscles off different strengths the fingers are a different lengths.

And.

It turns out actually that all but those are all there for a reason.

It turns out actually that all those are all there for a reason.

And.

Okay, so I'm gonna I'm gonna get quite a long answer to this question because it's because I have to unpack this question and then and then answer the unpacked version. Um, so

And.

So making it making it.

So making making it.

uh, first of all, uh,

The hand and for them because most of that most of the accurate just like the human hands.

The hand.

For them because most of that most of the accurate just like the human how deep the.

Most of that to control your hand or actually primarily in your forum.

The muscles of the control your hand or actually primarily in your forum.

Optimists hand, and forearm isn't it.

The optimists handled for them isn't it.

It'll be difficult engineering challenge.

It's I'd say it's that.

I have nothing but great things to say about Samsung. There are an amazing company um and Samsung is worth noting. Um does manufacturer um our AI for computer and does a great job doing that. Um, so

It's I'd say it's that.

um,

More difficult than the rest of from electromechanical standpoint, the forum in hand, it's more difficult than the entire rest of the robot.

More difficult than the rest of from electromechanical standpoint, the forum in hand, it's more.

Scope than the entire rest of the robot.

So, but really in order to have it.

So it really in order to have it.

A useful generalized robot.

A useful generalized grow but you do need this you need incredible hand.

You do need this you need incredible hand.

now with with the the AI 5 um and here's I I need to make a point of clarification relative to some comments. I made publicly uh before which is we're actually going to focus both tsmc and Samsung initially on AI 5,

And they need the real world AI.

And then you need the real World AI.

so, um,

And you need to built to scale up that production.

And you need to go to scale up that production.

The AI5 chip designed by Tesla is

But to have it be relevant because it's not relevant if it's just a few hundred robots.

I think it's an amazing design.

So to have it be relevant because it's not relevant if it's just a few hundred robots.

Um,

But so you need to be able to make.

I have spent almost every weekend for the last.

But so you need to be able to make.

Autonomous robots at volumes comparable to vehicles.

Last few months, uh, with the chip design team.

Working on, AI 5.

If not significantly higher.

and,

So I'm trying to get a million of something you per year My attorney Mcmillian Artemis robust for your debt.

I don't hand out. Praise

So I'm trying to make a million or something per year, turning like a million autonomous robots for your debt.

Elon Musk: By some metrics, the AI-5 chip will be 40 times better than the AI-4 chip. Not 40%, 40 times. We have a detailed understanding of the entire software and hardware stack, so we're designing the hardware to address all of the pain points in software. I don't think there really is anyone that's doing this, the entire stack, all the way through real-world, calibrating against the real world where you've got cars and robots in real world. We know what the chip needs to do, and just as importantly, we know what the chip doesn't need to do. To give you some examples here, with the AI-5, we deleted the legacy GPU or the traditional GPU, which is in AI-4. AI-5 does not have it; we just deleted the legacy GPU because it basically is a GPU. We also deleted the image signal processor.

Manufacturing challenges immense considering that the supply chain doesn't exist.

Easily. But I have to say that I think the Tesla trip team is really designing an incredible.

Manufacturing challenges immense considering that the supply chain doesn't exist.

Trip here. This is by some metrics.

Vaibhav Taneja: As discussed before, this business has a bigger impact from tariffs as measured by percentage of COGS, since currently all sales procured are from China, while we're still working on other alternatives. However, as the ramp of Megafactory Shanghai is happening, this is helping us avoid tariffs because we are using this factory to supply the non-US demand. Like Elon said, grid-scale storage, the only way we can get to electricity fastest is by using storage. The other thing to keep in mind is we are seeing headwinds in this business given the increase in competition and tariffs. The total tariff impacts for Q3 for both businesses was in excess of $400 million, generally split evenly between them. Services and other demonstrated a marked improvement sequentially. This was a function of improvements primarily in our insurance and service center businesses.

So with with cars, you've got an existing supply chain with computers, you've got an existing supply chain with.

The AI 5 chip will be 40 times better.

With a humanoid robot there is no supply chain.

Than the AI 4 chip not 40% 40 times.

With a humanoid robot there is no supply chain.

So in order to manufacture that has actually has to be.

Barry vertically integrated.

Very vertically integrated and.

And manufacturer.

Um, because we have a detailed understanding of the entire software and hardware stack, we're designing the hardware to, uh,

And manufacturer.

Hum very deep into the supply chain manufacture the parts internally because there just is no supply chain.

Address all of the pain points in software.

Very deep into the supply chain manufacture the parts internally.

It just has no supply chain.

So this is supposed to be kind of thing where I'm like if I put myself in a position of a startup trying to make and humanoid robot.

Uh, so I don't think there really is anyone that's doing this. I think the entire stack, um, all the way through the real world, uh,

So this is supposed to be kind of thing where I'm like if I put myself in the position of a start up trying to make a humanoid robot I'm like I don't know how to do it without.

You know, calibrating against the real world where you've got cars and robots in the real world, that.

I don't know how to do it without.

And an immense amount of manufacturing technology.

Like we know what the chip needs to do, and we know, just as importantly, what the chip does not need to do.

<unk>.

So.

That's why I think like Tesla is an almost unique I think I think unique position when you consider manufacturing technology scaling.

um, you know, sort of

That's why I think like Tesla is an almost unique I think I think unique position when you consider manufacturing technology scaling.

If you have some examples here, um, with the AI, we deleted the, um,

Real World AI.

And the and a truly dextrose hand.

And the <unk>.

Truly dextrose hand.

The Legacy GPU, or that, or the traditional GPU, which is it? It's in AI4.

Those are the generally the things that I'm missing when you read about other robots that they just don't have those three things.

Those are generally the things that I'm missing when you read about other robots that they just don't have those three things.

Vaibhav Taneja: Note that while small, our robotaxi costs are included within services and other along with our other businesses like paid supercharging, used car, parts, and merchandise sales, etc. Our operating expenses increased sequentially. The largest increase included in restructuring and other related to certain actions undertaken to reduce costs and improve efficiency through convergence of our AI chip design efforts. Additionally, we incurred legal expenses related to proceedings in certain legal cases, as well as incremental costs incurred in preparation for our shareholder meeting. Such costs are recorded within SG&A. Further, our employee-related spend is increasing, especially in R&D, as we have recently granted various performance-based equity awards to employees working on AI initiatives. Therefore, such spend will continue to increase going forward.

Um, but AI 5 does not have, um,

So I think we can.

So I think we can achieve.

We just, just, just delete it. The Legacy GPU, because it's basically a GPU.

But all those things those three things with an immense amount of work.

All those things those three things with an immense amount of work.

Elon Musk: This is a long list of deletions that are very important. As a result of these deletions, we can actually fit AI-5 in a half-radical and with good margin for the traces from the memory to the Tesla chip accelerators, the ARM CPU cores, and the PCI sort of the blocks. This is a beautiful chip. I've poured so much life energy into this chip personally. I'm confident this is going to be a winner, next level. It makes sense to have both Samsung and TSMC focus on AI-5. Even technically, the Samsung fab has slightly more advanced equipment than the TSMC fab. These will both be made in the U.S., one TSMC in Arizona, Samsung in Texas. We're going to make, starting off just to be confident of having our goal, explicit goal is to have an oversupply of AI-5 chips.

Uh, we also deleted the image signal processor. Um, and

And and that that is that is the game plan.

uh,

And and that that is that is the game plan.

This was like a long list of.

So.

Yeah My like my.

No fundamental concern with regard to.

How much voting control I Havent Tesla is if I go ahead and build this newest robot army can I just be outfit at some point in the future.

Actually, of the deletions that are very important. Um, as a result of these deletions, we can actually set AI 5 in a half radical.

How much voting control I Havent Tesla is.

Go ahead and build this newest robot army can I just be outfit at some point in the future.

That's my biggest concern if I could.

That was really the only thing I'm trying to address with with this.

Um, and with with, with good margin for the traces, from the memory to the the, the trip that that Tesla trip accelerators. Um,

What's called compensation, but it's not like I'm going to spend the money. It's just.

The arm, the ARM CPU course.

So what's called compensation, but it's not like I'm Gonna go spend the money it's just.

um,

and um,

So if we build this robot army.

And the PCI X sort of, uh,

Do I have at least.

Do I have at least a.

The PCI blocks. So, uh,

A strong influence over that robot.

Wrong influence over that robot not cannot control, but a strong influence.

This, this is a beautiful chair.

Trunk control, but a strong influence that's that's what it comes down to in a nutshell.

Um,

That's what it comes down to in a nutshell.

Like I don't feel comfortable holding it robot on me if I don't have at least a strong influence.

I don't feel comfortable holding it robot on me if I don't have at least a strong influence.

Great. Thank you and we've already covered robo taxi expansion.

Great. Thank you.

uh, I've heard so much life energy into this chip personally. Um, and I'm I'm confident this will be this is going to be a winner next level. Um so it makes sense to have

We've already covered robo taxi expansion.

Both Samsung and tsmc.

Unfortunately, the question after that is another future product questions. So we're going to have to skip that.

Focus on AI 5. Um,

And so even, like, technically, the...

The next one now is can you update us on the $16 $5 billion, Samsung chip deal and Taylor given the importance of semiconductors to autonomy and Tesla's AI driven future. What gives you confidence Samsung can fulfill AI six at Tesla timelines and achieve relatively better yields and cost versus T. S. T. S. M D.

um,

And, uh, but, but it's, uh,

Okay, So I'm going to get quite a long answer to those question.

Cause cause unpack those question and then and then answer the Unpacked version.

Because it's because unpack those question and then and then answer the Unpacked version.

If we're going to make progress, we need to be competent in achieving our goal. Our explicit goal is to have an oversupply of AI trip capacity.

Elon Musk: If we have too many AI-5 chips for the cars and robots, we can always put them in the data center. We already use AI-4 for training in our data center. We use a combination of AI-4 and NVIDIA hardware. We're not about to replace NVIDIA, to be clear, but we do use both in combination, AI-4 and NVIDIA hardware. The AI-5 excess production, we can always put in our data centers. NVIDIA keeps improving. The challenge that they have is that they've got to satisfy a large range or a lot of requirements from a lot of customers. Tesla only has to satisfy requirements from one customer, Tesla. That makes the design job radically easier and means we can delete a lot of complexity from the chip. I can't emphasize how important this is.

um,

So.

First of all.

I have nothing but great things to say about Samsung there.

Because if if we we have too many apps, I have chips for the cars and and and robots. We we can always put them in the data center.

Amazing company.

An amazing company.

So, we already use AI for.

And Samsung it was worth noting.

This manufacturer.

Manufacturer.

Our AI for computer and does a great job doing that.

Uh, for for training and in our data center. So we use a combination of AI for and, uh, Nvidia Hardware. Um,

So.

so, um,

So.

Now with the.

The <unk> five mm and heroes.

The <unk> five mm.

Travis Axelrod: Thank you very much, Vaibhav. Now, let's go to investor questions. From say.com, the first question is, "What are the latest robotaxi metrics? Fleet size, cumulative miles, rides completed, intervention rates, and when will safety drivers be removed? What are the obstacles still preventing unsupervised FSD from being deployed to customer vehicles?

And here as I.

To make a point of clarification relative to some comments I made publicly.

Need to make a point of clarification relative to some comments I made publicly.

Before which is we're actually going to focus both TSMC and Samsung initially on AI five.

We're not about to replace Nvidia, to be clear. But we do use both in combination: AI4 and video hardware, and the AI5 excess production. We can always put it in our data centers.

Um,

So.

The <unk> five chip designed by Tesla is.

Yeah, Nvidia keeps. It keeps improving um, that the challenge that they have is that they've got to satisfy a large range of a lot of requirements from a lot of customers.

I think it's an amazing design.

Elon Musk: I'll start off with that, and then Ashok can elaborate. But we are expecting to have no safety drivers in at least large parts of Austin by the end of this year. So within a few months, we expect to have no safety drivers at all, at least in parts of Austin. We're obviously being very cautious about the deployment. So our goal is to actually be paranoid about deployment because, obviously, even one accident will be front-page headline news worldwide. So it's better for us to take a cautious approach here. But we do expect to have no safety drivers in the car in Austin within a few months. I think that's perhaps the most important data point. And then we do expect to be operating robotaxi in, I think, about eight to 10 metro areas by the end of the year. It depends on various regulatory approvals.

But Tesla only has to satisfy requirements from one customer, but Tesla.

I have spent almost every weekend for last.

Last few months.

Last few months with the chip design team.

With the chip design team.

Working on AI five.

Elon Musk: When you look at the various logic blocks in the chip, as you increase the number of logic blocks, you also increase the interconnections between the logic blocks. You can think of it like there's just highways, like how many highways do you need to connect the various parts of the chip? Especially if you're not sure how much data is going to go between each logic block on the chip, then you kind of end up having giant highways going all over the place. It becomes an almost impossibly difficult design problem. NVIDIA has done an amazing job of dealing with almost an impossibly difficult set of requirements. In our case, we're going for radical simplicity.

And.

That makes the design job radically easier and means we can delete a lot of complexity from the joke, right? Like, I can't emphasize how important this is.

And.

I don't hand out praise.

Easily, but I have to say that I think I think that the.

Easily, but I have to say that I think I think the <unk>.

Chip team has really.

Hello Chip team has really.

Designing an incredible trip here this is by some metrics.

Designing an incredible Chuck here this is by some metrics.

Um, so like when you look at the various logic blocks in the chip, as you increase the number of logic blocks, you also increase the, uh, interconnections between logic blocks. So you can think of it like,

Yeah, if I have chip will be 40 times better than the AI fortunate not 40% 40 times.

Yeah, five chip will be 40 times better than the AI fortunate not 40% 40 times.

Because we have a detailed understanding of the entire software and hardware stack. So we're designing the hardware too.

Address all of the pain points in software.

<unk>.

Address all of the pain points in software.

So I don't think that there really isn't anyone that's doing this I think the entire stack.

Um, you know, just highways like, how many highways do you need to connect the various parts of the chip? Um, and especially if you're not sure how much data is going to go between each, you know, logic block on the chip, then you kind of end up having giant highways going all over the place. Um, it's a very, like, it becomes almost an impossibly difficult design problem. And Nvidia has done an amazing job of dealing with almost an impossibly difficult set of requirements. Um,

All the way through real World.

Elon Musk: The net effect is that I think AI-5 will be the best performance per watt, maybe by a factor of two or three, and the best performance per dollar for AI, maybe by a factor of 10. The proof's in the pudding. Obviously, we need to actually get this chip made and made at scale. That's what it looks like.

All the way through real World.

But in our case, we were going for radical simplicity.

Calibrating against the real World, where you've got cars and robust real world, but.

Um, and the net effect is that.

Calibrating against the real World, where you've got cars and robust real world.

I think AI 5 will be

We know what the chip needs to do and we know what.

Like we know what the chip needs to do and we know what.

The best performance per watt. Um,

Just as importantly, we know what the chip does it doesn't need to do.

Elon Musk: But you can actually think most of our regulatory applications are online. You can kind of see them because they're public information. But we expect to be operating in Nevada, Florida, and Arizona by the end of the year. Ashok?

Sort of.

Maybe by a factor of 2 or 3, um, and the best performance per dollar for AI maybe by a factor of 10.

To give you some examples here.

Maybe some examples here.

With a five.

so, you know, that's

With the <unk> five.

We deleted b.

The legacy GPU or that or the traditional GPU, which is it's in AI for.

The legacy GPU audit or the traditional GPU, which is at its in AI for.

Uh, you know what, we'll have to—the proof's in the pudding. So obviously, we need to actually get this trip made, um, and make it scale.

uh, but

Travis Axelrod: Yeah. We continue to operate our fleet in Austin without anyone in the driver's seat. We have covered more than a quarter million miles with that. And then in the Bay Area, where we still have a person in the driver's seat because of the regulations, we cross more than a million miles. We continue to see that the robotaxi fleet works really well. Customers are really happy, and there's no notable issues. On the customer side, customers have used FSD Supervised for a total of 6 billion miles as of yesterday. So that's like a big milestone. Overall, the safety continues to be very good. As Elon mentioned, we are on track to remove the person from inside the car altogether, starting with Austin. Great. The next question is, "What is the demand and backlog for Megapack, Powerwall, solar, or energy storage systems?

Ashok Elluswamy: Yeah. We continue to operate our fleet in Austin without anyone in the driver's seat. We have covered more than a quarter million miles with that. And then in the Bay Area, where we still have a person in the driver's seat because of the regulations, we cross more than a million miles. We continue to see that the robotaxi fleet works really well. Customers are really happy, and there's no notable issues. On the customer side, customers have used FSD Supervised for a total of 6 billion miles as of yesterday. So that's like a big milestone. Overall, the safety continues to be very good. As Elon mentioned, we are on track to remove the person from inside the car altogether, starting with Austin.

But AI five does not have.

that's what it looks like.

It.

Travis Axelrod: Great. Thank you, Elon. We've already covered unsupervised FSD. The next question is, instead of trying to replace hardware three with hardware four, why not give an equal incentive to trade in for a new vehicle?

It does not have.

We just just to leave it that the legacy GPU because it basically is a GPU.

Great. Thank you, Elon.

We just just to leave it that the legacy GP because it basically is a GPU.

So.

We also believe that the image signal processor.

And.

Vaibhav Taneja: Yeah. We've not completely given up on Hardware 3. However, over the last year, we've offered the customers the option to transfer Full Self-Driving to their new vehicle, which at times we've been running some promotions. If they got Full Self-Driving, they can get better preferential rates. We've been definitely taking care of this. We do want to solve autonomy first, and then we'll come back with a way to take care of these customers. These customers are very important. They were the early adopters. For what it's worth, my daily commuter is a Hardware 3 car, which I use Full Self-Driving on a daily basis. We will definitely take care of you guys.

Uh, we've already covered, uh, unsupervised, FSD. Um, so the next question is, instead of trying to replace Hardware 3 with Hardware 4, why not give an equal incentive to trade in for a new vehicle?

This feels like a long list of.

This feels like a long list of actually of deletions that are very important.

Actually of deletions that are very important.

Yeah, we've not completely given up on Hardware 3.

However,

As a result of these deletions, we can actually put out five.

And Ah have radical.

And.

With with with good margin for the traces from the memory too that the trip.

Tesla trip accelerators.

The army the army CPU cores.

Travis Axelrod: Great. The next question is, "What is the demand and backlog for Megapack, Powerwall, solar, or energy storage systems?

And.

And the PCI.

Sort of.

The PCI blocks so.

This this is a beautiful job.

over the last year, we've offered the customers, the option to transfer, FSD to their new wake up with which, at times, we've been running some promotions. If, if they got FSD, they can get better. Preferential rates. So we've been definitely taking care of this, but we do want to solve autonomy first and then we'll come back with a way to take care of these customers. These customers are very important. They were the early adopters for what. It's worth my daily commuter is the hardware 3 car which I use FSD on a daily basis, so we will definitely take care of you guys.

Travis Axelrod: Great. Thank you.

Travis Axelrod: With the current AI boom, is Tesla planning to supply power to other hyperscalers?

Ashok Elluswamy: Once the FSD version 14 release series is fully done, we are planning on working on a FSD version 14 Lite version for hardware three, probably expected in Q2 next year.

I've heard so much life energy cause shall personally.

I've heard so much life energy into this shall personally.

And I'm confident this will be this is going to be a winter next level.

Ashok Elluswamy: Thanks. Demand for Megapack and Powerwall continues to be really strong into next year. We received very strong positive customer feedback on our Megablock product, which will begin shipping next year out of Houston. And we're seeing remarkable growth in the demand for AI and data center applications as hyperscalers and utilities have seen the versatility of the Megapack product to increase reliability and relieve grid constraints, as Elon was talking about. We've also seen a surge in residential solar demand in the US due to policy changes, which we expect to continue into the first half of 2026 as we introduce a new solar lease product. And we also began production of our Tesla residential solar panel in our Buffalo factory, and we will be shipping that to customers starting Q1. The panel has industry-leading aesthetics and shape performance and demonstrates our continued commitment to US manufacturing.

Lars Moravy: Thanks. Demand for Megapack and Powerwall continues to be really strong into next year. We received very strong positive customer feedback on our Megablock product, which will begin shipping next year out of Houston. And we're seeing remarkable growth in the demand for AI and data center applications as hyperscalers and utilities have seen the versatility of the Megapack product to increase reliability and relieve grid constraints, as Elon was talking about. We've also seen a surge in residential solar demand in the US due to policy changes, which we expect to continue into the first half of 2026 as we introduce a new solar lease product. And we also began production of our Tesla residential solar panel in our Buffalo factory, and we will be shipping that to customers starting Q1. The panel has industry-leading aesthetics and shape performance and demonstrates our continued commitment to US manufacturing.

And I'm confident this will be this is going to be a winter next level.

So it makes sense to have.

Both Samsung and TSMC.

Great, and thank you, Edition. Um, once the V4 release series is fully done, we are planning on working on a V14 Lite version for Hardware 3, probably expected in Q2 next year.

Both Samsung and TSMC.

Travis Axelrod: Awesome. Thanks, Ashok. All righty. Our final question from Say is, how long until we see self-driving Tesla semi-trucks? Could you see this technology replacing trains?

On <unk> five mm.

Focus on <unk> five.

Awesome. Thanks a joke.

So you mean like the technically the.

Yeah. So you mean like the technically the.

Samsung has slightly more advanced equipment, then they TSMC fab will be both be made in the.

Samsung has.

Yes.

Slightly more advanced equipment, then they TSMC fab people would both be made in <unk>.

Lars Moravy: Yeah. I guess I'll start with that in terms of the semi-production plan and schedule. The factory is going on schedule. We've completed the building and are installing the equipment now. We've got our fleet of validation trucks driving on the road. We'll have larger builds towards the end of this year and then our first online builds in the first part of next year, ramping into the Q2 timing with real volume coming in the back half of the year. That's going quite well. That's the first step, obviously, getting autonomous trucks on the road. In terms of trains, you know they're really great for long point-to-point deliveries. They're super efficient. That last mile, the load unload can be better served for shorter distances with autonomous semis. That would be great.

The U S.

All righty. Um, our final question from say is, uh, how long until we see self-driving Tesla semi trucks? Uh, and could you see this technology replacing trains?

In the U S.

TSMC in Arizona, Samsung in Texas.

And a bit but it's.

We're gonna make starting off just to be competent of having.

We're gonna make starting off just to be confident of having.

I called explicit goal was to have an oversupply of AI pipe chips.

Alcoholics explicit goal is to have an oversupply of AI pipe chips.

Yeah.

Because if we have too many F <unk> chips for the cars.

Because if we have too many at <unk> chips for the cars.

And robots, we can always put them in the data center.

And robots, we can always put them in a data center.

So we already use AI for.

Or for training.

Our data center, so we use a combination of airport and.

Data center, so we use a combination of airport and.

Travis Axelrod: Great. Thank you, Mike. Unfortunately, the next question is related to future products. This is not the appropriate venue to cover that, so we're going to have to skip it. The question after that is, "What are the present challenges in bringing Optimus to market, considering app control software, engineering hardware, training general mobility models, training task-specific models, training voice models, implementing manufacturing, and establishing supply chains?

Nvidia hardware.

In video hardware.

Yeah so I guess I'll start with that in terms of the semi uh production plan and schedule. So the factory is going on schedule. We've you know completed the building and are installing the equipment. Now um we've got our Fleet of validation trucks driving on the road. We'll have larger builds towards the end of this year and then our first online builds in the first part of next year, uh, ramping into, you know, the Q2 timing with real volume coming in the back half of the year. Um, so that's going quite well. And, and that's the first step obviously getting autonomous. Uh, trucks on the road. Um, in terms of trains, you know, they're really great for Long Point to Point, deliveries, they're super efficient. But you know, that last mile the load unload. Uh,

So.

But we're not about to replace and video to be clear, but we do use both.

Lars Moravy: We do expect that to probably shift in as we, you know really, as Elon said, change the way transportation is considered. We're looking forward to that timeline. Ashok, I know you can take the Full Self-Driving part.

We're not about to replace and video to be clear, but we do use both.

In combination.

And for video hardware.

Therefore in the video hardware.

And there are five excess production, we can always put it in our data centers.

Had changed the way transportation is considered.

And there are five excess production, we can always put in our data centers.

Ashok Elluswamy: Currently, the team is super focused on solving for passenger vehicle autonomy. That said, the same technology will extend quite easily to the semi-truck once we have a little bit of data from the semi-trucks.

And really it keeps it keeps improving.

The challenge that they have is that they've got to satisfy a large range.

But the challenge that they have is that they've got to satisfy a large range.

A lot of requirements from a lot of customers.

Elon Musk: Yeah. I mean, bringing Optimus to market is an incredibly difficult task, to be clear. It's not like some walk in the park at some point. I mean, actually, technically, Optimus can walk in the park right now. And we do have Optimus robots that walk around our offices at our engineering headquarters in Palo Alto, California, basically 24 hours a day, seven days a week. So any visitors that come by, you can stop one of the Optimus robots and ask it to take you somewhere, and it'll literally take you to that meeting room or that location in the building. So I don't want to downplay the difficulty of Optimus. It's an incredibly difficult thing. Especially, it's difficult to create a hand that is as dexterous and capable as the human hand, which is an incredible, the human hand is an incredible thing.

Tessa only has to satisfy requirements from one customer a tesla.

Um, and so we're looking forward to that timeline and Ashok. I know you uh, can can take the the full self-driving part. Currently, the team is like super focused on uh solving for passenger vehicles. Autonomy that said the same technology will uh actually quite easily to the summit truck. Once we are a little bit of data from The Summit trucks.

Travis Axelrod: Great. We will move over to analyst questions. The first question comes from Emmanuel at Wolfe. Emmanuel, please go ahead and unmute yourself.

That makes the designs are all about it.

That makes the design job radically easier it means we can delete.

Easier and means we can delete.

A lot of complexity from the truth.

A lot of complexity from the chip I can't emphasize how important this is.

Like I kind of set top boxes.

So like when you look at the various logic blocks in the chip.

Great. Uh, and now we will move over to analyst questions. The first question comes from Emmanuel at Wolfe. Emmanuel, please go ahead and unmute yourself.

So like when you look at the various logic blocks in the chip.

[Analyst 1]: Great. Thanks so much. Hi, everybody. Elon, you talked about expanding production of vehicles as fast as possible now that you have confidence in the unsupervised autonomy. How should we think about that in the context of your existing capacity of 3 million units? Is that where you're hoping to get volume to? What sort of timeline are we talking about? Would this require some level of boosting or incentivizing demand? Would this basically be prioritizing volume over near-term profitability given the longer-term opportunity?

As you increase the number of logic blocks, you will see an increase.

As you increase the number of logic blocks, you will see increasing the.

Interconnections between logic blocks, which you can think of it like.

Interconnections between logic blocks. So that you can think of it like.

They're just highways like how many highways do you need to connect the various parts of the chip.

And especially if you're not sure.

And especially if you're not sure how.

How much data is going to go between.

Each.

Logic block on the Chip then you kind of end up having tried highways going all over the place.

It's a very it's like it becomes almost impossibly difficult.

It's a very it's like it becomes an almost impossibly difficult.

Design problem and Nvidia has done an amazing job of dealing with almost impossibly difficult set of requirements.

Designed problem and Nvidia has done an amazing job of dealing with almost impossibly difficult set of requirements.

But in our case we.

Elon Musk: The more you study the human hand, the more incredible you realize the human hand is and why you need four fingers on the thumb, why the fingers have certain degrees of freedom, why the various muscles are of different strengths, the fingers are of different lengths. And so making the hand and forearm, because most of the actuator, just like the human hand, the muscles that control your hand are actually primarily in your forearm, the Optimus hand and forearm is an incredibly difficult engineering challenge. I'd say it's more difficult than the rest of, from an electromechanical standpoint, the forearm and hand are more difficult than the entire rest of the robot. So in order to have a useful generalized robot, you do need this. You do need an incredible hand.

Great for radical simplicity.

We're paying for radical simplicity.

And the net effect is that.

Great. Uh, thanks so much. Uh, hi everybody. Um, so Elon, you talked about, um, expanding production of vehicles as fast as possible now that you have confidence in the unsupervised autonomy. Um, how should we think about that in the context of your existing capacity of 3 million units? Um, is that, um, where you're hoping to get volume to? What sort of timeline are we talking about? And, uh, would this require, you know, some level of, uh, boosting or incentivizing demand? Like, would this basically be prioritizing volume over near-term profitability given the longer-term opportunity?

Elon Musk: Our capacity isn't quite 3 million, but it will be 3 million at some point. Aspirationally, it could be 3 million within, we could probably hit an annualized rate of 3 million within 24 months, I think. Maybe less than 24 months. Bearing in mind, there's an entire supply chain, a vast supply chain that's got to also move in tandem with that. We are going to expand production as fast as we can and as fast as our suppliers can sort of keep up with it. We've got to think about where we build incremental factories beyond that. The single biggest expansion in production will be the Cybercab, which starts production in Q2 next year. That's really a vehicle that's optimized for full autonomy.

I think and five will be.

I think <unk> five will be the.

The best performance per watt.

Well, uh, the capacity isn't quite 3 million. Um,

The best performance per watt.

Maybe by a factor of two or three.

And the best performance per dollar for AI, maybe by a factor of 10.

But, uh, it will be $3 million at some point. Um,

So you know that's.

That's.

You know what we'll have to the proofs in the pudding. So obviously, we need to actually get this chip made.

You know what we'll have to the proofs in the pudding. So obviously, we need to actually get the ship made.

You know, aspirationally you know, it could be 3 million within, we could probably hit an annualized rate of 3 million within 24 months. I think, um, maybe less than 24 months um

And made it scale.

But.

That's what it looks like.

Bearing in mind like this. Like, there's an entire like supply chain, like, a vast supply chain, that's got to also move in tandem with that. Um,

Great. Thank you Elon.

Already covered unsupervised F S D.

Already covered unsupervised FSD and.

So the next question is instead of trying to replace hardware three with hardware for why not give an equal incentive to trading for a new vehicle.

so but we're going to, we're going to expand, uh, production as fast as, as, as as as we can, um, as fast as our suppliers can can, um,

can sort of keep up with it, um,

Yeah.

Not completely given up on hardware.

Our board last year, we offer the customers the option to transfer SSD to their new Waco.

And then we're going to think about, uh, where do we build incremental factories? Uh, beyond that?

Our or last year, we offer the customers the option to transfer that says D to their new Waco.

Um,

Which at times had been running some promotions if if they got FSD. They can get better preferential rates. So we've been definitely taken care of this but we do want to solid taught me first and then we'll come back with a way to take care of these customers. These customers are very important they were the early adopters for what it's worth.

Elon Musk: Then you need the real-world AI, and you need to be able to scale up that production to have it be relevant, because it's not relevant if it's just a few hundred robots. So you need to be able to make Optimus robots at volumes comparable to vehicles, if not significantly higher. So if you're trying to make a million or something per year, trying to make a million Optimus robots per year, that manufacturing challenge is immense, considering that the supply chain doesn't exist. So with cars, you've got an existing supply chain. With computers, you've got an existing supply chain. With a humanoid robot, there is no supply chain. So in order to manufacture that, Tesla actually has to be very vertically integrated and manufacture very deep into the supply chain, manufacture the parts internally because there just is no supply chain.

Which at times had been running some promotions if they got FSD. They can get better preferential rates. So we've been definitely taking care of this but we do want to solve taught me first and then we'll come back with a way to take care of these customers. These customers are very important they were the early adopters for what it's worth.

Elon Musk: It, in fact, does not have a steering wheel or pedals and is really an engineering optimization on minimizing cost per mile, fully considered cost per mile of operation. For our other vehicles, they still have a little bit of the horseless carriage thing going on where, obviously, if you've got steering wheels and pedals and you're designing a car that people might want to go, you know, very do fast acceleration and tight cornering, like high-performance cars, then you're going to design a different car than one that is optimized for a comfortable ride, but doesn't expect to go past sort of 85 or 90 miles an hour and is just aiming for a gentle ride the whole time. That's what Cybercab is. Do I think we'll sacrifice margins? I don't think so. I think the demand will be pretty nutty.

Like the single biggest expansion in production will be the the Cyber cap which starts production um in Q2 next year. Um that's uh, that's really a vehicle that's optimized for full autonomy. Uh, it in fact does not have a steering wheel or pedals um, and is really um,

And engineering optimization on minimizing cost for, like, fully considered cost per mile of operation.

um,

My daily commute is a hardware CCAR, which I use SSD on a daily basis. So we will definitely take care of you guys.

My daily commute is the hardware CCAR, which I use SSD on a daily basis. So we will definitely take care of you guys.

So that's, you know, for for the other, for the, for our other vehicles, there's still, they still have a little bit of the Horseless Carriage thing going on. Where, you know, obviously, you've got if you're still, if you've got steering wheels and pedals and, um,

Great. Thank you.

Once the 14.

It's fully done.

Our planning on working on our 14 light version photography, probably expected in Q2 next year.

Our planning on working on our repo in light version 400 feet, probably expected in Q2 next year.

Yeah.

Awesome. Thank you Chuck.

Awesome. Thanks Chuck.

Alrighty. Our final question from say is how long until we see self driving Tesla semi trucks and <unk>.

Already our final question from say is how long until we see self driving Tesla semi trucks and could you see this technology, replacing trains.

And and you're designing a car that uh people might want to go, you know, very do right, past acceleration and tight cornering like high performance cars, you know the cars then you're going to design a different car than 1 that is optimized for accountable ride. Um, but doesn't expect to go, you know, past sort of 85 or 90 miles an hour, um, and it's it's just aiming for a gentle ride the whole time. That's what cyber cap is. Um,

Could you see this technology, replacing trains.

so, um,

Yeah, So I guess I'll start with that in terms of the semi a production plan and schedule. So the factory is it's going on schedule we've.

Yeah, So I guess I'll start with that in terms of the semi hmm production plan and schedule. So the factory is it's going on schedule we've.

yeah, so so it's

Elon Musk: So this is the kind of thing where I'm like, if I put myself in the position of a startup trying to make a humanoid robot, I'm like, "I don't know how to do it without an immense amount of manufacturing technology." So that's why I think Tesla's in almost a unique position when you consider manufacturing technology scaling, real-world AI, and a truly dexterous hand. Those are generally the things that are missing when you read about other robots that just don't have those three things. So I think we can achieve all those things, those three things, with an immense amount of work. And that is the game plan. So my fundamental concern with regard to how much voting control I have in Tesla is, if I go ahead and build this enormous robot army, can I just be ousted at some point in the future?

Elon Musk: Here's the killer app, really, what it comes down to is can you text it? Can you text while you're in the car? If you tell someone, "Yes, the car is now so good," you can be on your phone and text the entire time while you're in the car. Anyone who can buy the car will buy the car, end of story. That's what everybody wants to do. In fact, not everybody wants to do it. They do do that. The reason you've seen that there's been an uptick in accidents pretty much worldwide is because people are texting and driving. Autopilot actually dramatically improves the safety here because if somebody's looking down at their phone, they're not driving very well. That's really the game changer.

Completed the building and are installing equipment now.

Do I think we'll sacrifice margins? I don't think so. I think the demand will be pretty nutty.

Completed the building and are installing equipment now.

We've got a fleet of validation trucks are driving on the road will have larger builds towards the end of this year and then our first online builds in the first part of next year ramping into Q2 timing with real volume coming in the back half of the year.

Um, but like, here's the killer app. Really, what it comes down to is, can you text it? Can you text while you're in the car?

And if you tell someone yes that the car is now so good. You can you can you can be on your phone.

uh,

So that's going quite well and that's the first step obviously getting autonomous trucks on the road.

and text the entire time while you're in the car.

So that's going quite well and that's the first step obviously getting autonomous trucks on the road.

It's

Anyone who can buy the car will buy the car.

In terms of the trains you know there really great for long point to point deliveries are superefficient, but you know that last mile to load unload.

And a and a story.

In terms of the trains you know there really great for long point to point deliveries are superefficient, but you know that last mile of the load unload.

um,

so,

Right.

Can be better served for and shorter distances with autonomous semis and that would be great and so we do expect that to probably shift and as we really change the way transportation is considered.

Right.

Can be better served for shorter distances with autonomous semis and that would be great and so we do expect that to probably shift and as we really as you know.

Um, that's what everybody wants to do, and in fact, not everyone wants to do it; they do do that. And that's why, in fact, the reason you've seen, like, there's been an uptick in accidents.

Pretty much worldwide, it's because people are texting and driving.

<unk> changed the way transportation is considered.

And so we're looking forward to that timeline and Ashoka I know you can can take the full self driving part currently the team is super focused on solving for passenger vehicles autonomy.

And so we're looking forward to that timeline and Ashoka I know you can can take the full self driving for.

Um, so, uh, Autopilot actually improves safety here because if someone is looking down at their phone, they are not driving very well.

Um,

The team is super focused on solving for passenger vehicles autonomy.

Elon Musk: You know, we do see, like at this point, I feel, you know, essentially 100% confident, I say not essentially, 100% confident that we can solve unsupervised Full Self-Driving at a safety level much greater than human. We have released 14.1. We have got a technology roadmap that is, I think, pretty amazing. We will be adding reasoning to the car. Our world simulator for reinforcement learning is pretty incredible. When you see the Tesla reality simulator, you cannot tell the difference between the video that is generated by the Tesla reality simulator and the actual video. It looks exactly the same. That allows us to have a very powerful reinforcement learning loop to further improve the Tesla AI. We are going to be increasing the parameter count by an order of magnitude. That is not in 14.1.

So, that's really the game changer.

The same technology even.

Same technology even.

and uh,

Suddenly quite easily.

Actually quite easily to its I'm not sure I'd want to be.

Once we have data from the semi trucks.

Data from the semi trucks.

You know, we see that, at this point, I feel.

Great.

Elon Musk: That's my biggest concern. That is really the only thing I'm trying to address with what's called compensation, but it's not like I'm going to go expend the money. It's just, if we build this robot army, do I have at least a strong influence over that robot army? Not control, but a strong influence. That's what it comes down to in a nutshell. I don't feel comfortable building that robot army if I don't have at least a strong influence.

Now we will move over to analyst questions. The first question comes from Emmanuel at Wolf Manuel. Please go ahead and them yourself.

And now we will move over to analyst questions.

First question comes from Emmanuel at Wolf.

You know, essentially 100% confident, I say not essentially 100% confident that we can set we that we can solve unsupervised. Full drive, full self-driving at a safety level.

Please go ahead and do them yourself.

Um, much greater than human.

um,

Great. Thanks, so much hi, everyone. So Elon you talked about expanding production of equal just fastest possible now that you have confidence in the unsupervised autonomy, how should we think about that in the context of your existing.

Great. Thanks, so much hi, everybody.

So Elon you talked about expanding production of vehicles as fast as possible now that you have confidence in the unsupervised autonomy, how should we think about that in the context of your existing capacity of 3 million units is that where you're hoping.

Adding reasoning to the car, uh, uh, our world simulator for SIM for reinforcement learning is.

The city of 3 million units is that where youre, hoping to get volume to what sort of timeline are we talking about and would.

Travis Axelrod: Great. Thank you. We've already covered robotaxi expansion. Unfortunately, the question after that is another future product question, so we're going to have to skip that. The next one, though, is, "Can you update us on the $16.5 billion Samsung chip deal in Taylor? Given the importance of semiconductors to autonomy in Tesla's AI-driven future, what gives you confidence Samsung can fulfill AI6 at Tesla's timelines and achieve relatively better yields and cost versus TSMC?

To get volume to what sort of timeline are we talking about and.

It's pretty incredible, like our, like, our when you see the Tesla reality simulator. Um, you can't tell a difference between.

Would this require some level of boosting or incentivizing the man like would just.

Would this require you know.

Some level of boosting or incentivizing the men like would just basically.

The video that's generated by the Tesla reality simulator and the actual video look exactly the same.

Basically be prioritizing volume over near term profitability, even the longer term opportunity.

Well, our capacity isn't quite 3 million.

But.

Elon Musk: There are also a number of other improvements to the AI that are quite radical. This car will feel like it is a living creature. That is how good the AI will get with the AI-4 computer. This is before AI-5. AI-5, like I said, is by some metrics 40 times better, which is safely a 10x improvement. It might almost be too much intelligence for a car. I do wonder, like how much intelligence should you have in a car? It might get bored, actually. One of the things I thought of, like, if we have got all these cars that maybe are bored, while they are sort of, if they are bored, we could actually have a giant distributed inference fleet and say, like, if they are not actively driving, let us just have a giant distributed inference fleet.

But.

It will be 3 million at some point.

Elon Musk: Okay. So I'm going to give quite a long answer to this question because I have to unpack this question and then answer the unpacked version. So first of all, I have nothing but great things to say about Samsung. They're an amazing company. And Samsung, it is worth noting, does manufacture our AI4 computer and does a great job doing that. So now, with the AI5, and here, I need to make a point of clarification relative to some comments I've made publicly before, which is we're actually going to focus both TSMC and Samsung initially on AI5. So the AI5 chip design by Tesla, I think it's an amazing design. I have spent almost every weekend for the last few months with the chip design team working on AI5.

Aspirational.

Um, so that that allows us to, um, have a very powerful reinforcement learning Loop. Uh, to further improve the Tesla AI. We're, we're going to be increasing the parameter count by an order of magnitude. Um that that's not important 14.1. Um there are also a number of other improvements to the AI. Just um

Aspirational.

It could be 3 million within <unk>.

Probably at an annualized rate of 3 million within 24 months I think maybe less than 24 months.

That, that are that are quite radical. Um,

Probably at an annualized rate of 3 million within 24 months, I think and maybe less than 24 months.

so it's, uh,

Bearing in mind like this like there's an entire like supply chain.

A vast supply chain, that's going to also move in tandem with that.

This call will feel like it is a living creature. That's how good the the AI will get. What the ai4 computer. This is before AI 5.

As supply chain, that's going to also move in tandem with that.

So I think.

So, but we're going to we're going to expand.

We're going to work.

We're going to expand our production as fastest as we can.

Production as fastest as we can.

And then, and then AI 5, like I said, is by some metrics 40 times better. Um, let's just say safely it's a 10x improvement.

Can and.

And as fast as our suppliers can can.

And as fast as our suppliers can you can.

And you can sort of keep up with it.

So it might only be too much intelligence for a car. I do wonder, like, how much intelligence should you have in a car? It might get bored.

um,

And then we're going to think about a world where do we build incremental factories.

And then we're going to think about where do we build incremental factories.

actually, um,

On bad debt.

That like.

The single biggest.

The single biggest expansion in production would be that the cyber cap, which started production in Q2 next year.

Production will be the cyber cap, which starts production in Q2 next year.

That's a that's really a vertical that's optimized for full autonomy.

In fact does not have a steering wheel or pedals.

and then 1 of the things, I thought like, well if we got all these cars that maybe a board. Well, why they're while they're sort of, if they are bored, we could actually have a giant distributed in influence Fleet and say like well if they're not actively driving, let's just have a giant distributed inference Fleet.

In fact does not have a steering wheel or pedals.

Elon Musk: At some point, if you have got like tens of millions of cars in the fleet, or maybe at some point, 100 million cars in the fleet, and let us say they had at that point, you know, I do not know, a kilowatt of inference capability of, you know, high-performance inference capability. That is 100 gigawatts of inference distributed with power and cooling taken with cooling and power conversion taken care of. That seems like a pretty significant asset.

um,

And it's really.

And engineering optimization on minimizing costs might be like fully considered cost per mile of operation.

And engineering optimization on minimizing cost per mile.

For like fully considered cost per mile of operation.

You know, at some point, if you've got like tens of millions of cars in the fleet, or maybe at some point 100 million cars in the fleet, um,

and um,

So that's.

So that's what's worked for other for other vehicles, they're still they still have a little bit of the horseless carriage thing going on with you. Obviously, you got but you still got steering wheels and pedals and.

For the other party.

Elon Musk: And I don't hand out praise easily, but I have to say that I think the Tesla chip team is really designing an incredible chip here. By some metrics, the AI5 chip will be 40 times better than the AI4 chip. Not 40%, 40 times. Because we have a detailed understanding of the entire software and hardware stack. So we're designing the hardware to address all of the pain points in software. So there really isn't anyone that's doing the entire stack all the way through real-world calibrating against the real world where you've got cars and robots in the real world. We know what the chip needs to do, and we know what, just as importantly, we know what the chip doesn't need to do. Sort of give you some examples here. With the AI5, we deleted the legacy GPU or the traditional GPU, which is in AI4.

For other vehicles, they're still they still have a little bit of the horseless carriage thing going on with you. Obviously, you got but you still got steering wheels and pedals and.

Let's say they had, at that point, you know? Uh,

Like, I don't know.

And you're driving a car that.

People might want to go up.

Uh, a kilowatt of influence capability of, you know, high performance in Ananse capability. That's 100 gigawatts of inference.

People might want to go from a very.

Very fast.

Very fast acceleration and tight cornering like high performance Court.

Past acceleration of tight cornering like high performance.

Distributed.

The cars, then you're going to design a different car than one that is optimized for a comfortable ride.

Cars, then you're going to design a different car than one that is optimized for a comfortable ride.

With power and cooling taking care of, with cooling and power conversion, taking care of.

so,

But doesn't expect to go past sort of 85 to 90 miles an hour.

That seems like a pretty significant asset.

Travis Axelrod: Great. Thanks, Elon. The next question comes from Adam Jonas from Morgan Stanley. Adam, please feel free to unmute yourself. Adam, go ahead and ask your question. Seems like we might be having some audio issues with Adam. We'll come back to you. The next question will then come from Dan Levy from Barclays.

And it's just Amy for a gentle ride the whole time, that's what cyber cafes.

And it's just aiming for a gentle right the whole time, that's what cyber cafes.

So.

Great. Thanks, Elon. Uh, the next question comes from Adam, uh, from Morgan Stanley. Adam, please feel free to unmute yourself.

Yeah. So it's.

Do I think we will sacrifice margins I don't think so I think the demand will be pretty nutty.

I do I think we will sacrifice margins I don't think so I think demand will be pretty nutty.

Uh, Adam, go ahead and ask your question.

Like here's the here's the killer App really what it comes down to is can your text can you tax while you're in the car.

And if your telephone yes. The cars now so good you can you can you can be on your phone.

It seems like we might be having some audio issues with Adam, so we'll come back to you. The next question will then come from Dan at Barclays.

Yeah.

Dan Levy: Hi. Good evening. Thank you for taking the question. Elon, I know that Tesla is really focused on with master plan for bringing AI into the physical world. I think we've seen over the past, you know, this willingness for Tesla to engage and go into new markets, new TAMs. When you think about the growth prospects, how do we define the areas that are really within Tesla's core competency versus where do you draw the line for markets or AI applications that are outside of Tesla's core competency?

And text the entire time, while you're in the car.

It's.

Hi, good evening. Thank you for, uh, taking, uh, the

It's <unk>.

Anyone who can buy the car will buy the car.

um,

And is it <unk>.

Sorry.

Is it mandatory.

So.

That's what everybody wants to do in fact, not everyone wants to if they do do that and that's why in fact, the reason you've seen like that theres been an uptick in accidents.

Elon Musk: But AI5 does not have, we just deleted the legacy GPU because it basically is a GPU. So we also deleted the image signal processor. And this is like a long list of deletions that are very important. As a result of these deletions, we can actually fit AI5 in a half reticle with good margin for the traces from the memory to the Tesla chip accelerators, the ARM CPU cores, and the PCI blocks. So this is a beautiful chip. I've poured so much life energy into this chip personally, and I'm confident this will be, this is going to be a winner next level. So it makes sense to have both Samsung and TSMC focus on AI5. So technically, the Samsung fab has slightly more advanced equipment than the TSMC fab. These will both be made in the US, one TSMC in Arizona, Samsung in Texas.

That's what everybody wants to do in fact, not everyone wants to if they do do that and that's why in fact, the reason you've seen like that theres been an uptick in accidents.

You know, and I know that Tesla's really focused on with Master Plan for bringing in AI into the physical world. And I think we've seen over the past,

You know, this willingness for Tesla to engage and go into new markets, new TAMs. So,

Sure.

Pretty much worldwide is because people are texting and driving.

So autopilot extra dramatically improves the safety here.

Because if somebody is looking down at their phone, they're not driving very well.

So that's that's really the game changer.

When you think about the growth prospects, how do we define the areas that are really within Tesla's core competency versus where do you draw the line for markets or AI applications that are outside of Tesla's core competency?

Elon Musk: Actually, I'm not sure what you mean by AI applications outside of Tesla's core competency. I mean, we kind of didn't have any of these core competencies when we started, you know. It's like we had zero core competencies, total competency of zero, actually. You can think of Tesla as like, I don't know, a dozen startups in one company. I've initiated every one of those startups. We didn't use to make battery packs, say stationary battery packs, but now we do. We make them for the home, make them for, you know, utility scale with Powerwall, Megapack. We created the Supercharging network globally. No one else has created a global Supercharging network. In fact, that North American Supercharging network is so good that basically, every other manufacturer in North America has converted to our standard and uses the Tesla Supercharging network.

And.

We do see like.

At this point I feel.

Um, actually, I'm not sure what you mean by applications outside of Tesla's core competency. Um,

At this point I feel.

Uh huh.

but uh,

You know essentially 100% confident I say not essentially 100% confident that we can say that we can solve unsupervised goldstrike full self driving a safety level.

Essentially 100% confident I say not essentially 100% confident that we can set that we can solve.

and we, we kind of

Bicycles Trikes full self driving at a safety level.

Much greater than human.

So we've released 14.1, we've got a technology roadmap, that's I think a pretty amazing, we'll be adding reasoning to the car.

We didn't have any of these core competencies when we started, you know? Um, so it's like, we had zero core competencies. Total competency of zero, actually. Um, so we, I mean, you can think of Tesla as like, I don't know, a dozen startups in one company. Um,

You know, and, uh, I've initiated every one of those startups. So,

it's uh,

Our wealth simulator for assembly for reinforcement learning is.

Our world simulator for Assembly for reinforcement learning is.

Yeah. We're getting used to making battery packs, say stationary battery packs, but now we do make them for the home. We make them for, you know, utility scale with.

It's pretty incredible.

Our mega pack.

It's pretty incredible.

When you say that that has a reality simulator.

uh,

When you say that that is a reality something later.

It's you can tell it on a screen.

The video that's generated by the Tesla reality simulator and but the actual video looks exactly the same.

The video that's generated by the Tesoro reality simulator and the <unk>.

Actual video looks exactly the same.

We created the Supercharger Network globally. I won't know anyone else who has created a global Supercharger Network. In fact, the North American Supercharger Network is so good at that, that basically...

So that allows us to.

Yeah, every other manufacturer in North America has converted to our standard.

Have a very powerful reinforcement learning loop.

You have a very powerful reinforcement learning loop.

Elon Musk: If it was so easy, why don't they just do it? The chip design team, we started that from scratch. The Tesla AI software team, we started from scratch. I literally just say, "Hey, we're going to start this thing." I post it on Twitter, now X, and then, you know, join us if you'd like to build it. In fact, Ashok was, I believe, the first person I interviewed for the Tesla Autopilot team, which we now call the Tesla AI Software Team, because it is the AI Software Team. It's core competencies created while you wait. Optimus at scale is the infinite money glitch. It's difficult to express the magnitude of, like, if you've got something that, like, if Optimus, I think, could probably achieve 5x the productivity of a person per year because it can operate 24/7. It doesn't even need to charge.

Further improve that tells the AI, we're gonna be increasingly parameter account by an order of magnitude.

To further improve that tells the AI, we're gonna be increasingly parameter account by an order of magnitude.

Um, and uses our Tesla Supercharger network. Um, but if it was so easy, why don't they just do it?

But that's not in 14.1.

Elon Musk: But we're going to make, starting off just to be confident of having our explicit goal is to have an oversupply of AI5 chips. Because if we have too many AI5 chips for the cars and robots, we can always put them in the data center. So we already use AI4 for training in our data center. So we use a combination of AI4 and NVIDIA hardware. So we're not about to replace NVIDIA, to be clear, but we do use both in combination, AI4 and NVIDIA hardware. And the AI5 excess production, we can always put in our data centers. NVIDIA keeps improving. The challenge that they have is that they've got to satisfy a lot of requirements from a lot of customers. But Tesla only has to satisfy requirements from one customer, but Tesla.

Um, and, uh, the chip design team.

That's not in 14.1, they're also a number of other improvements to the AI. It just.

There are also a number of other improvements to the AI. It just.

um,

Yeah.

That that are that are quite radical.

That better that are quite radical.

So it's.

I literally just said, "Hey, we're going to start this thing." I posted on Twitter now, X, and then.

This call will feel like it is a living creature. That's how good people he and I will get with the EF all computer with this before a high five.

This call will feel like it is a living creature that is how good he and I will get with the asphalt computer before AI five.

And then and then AI five like I said is by some metrics 40 40 times better.

And then like I said is by some metrics 40, 40 times better.

You know, join us if you'd like to build it. In fact, Ashok was, I believe, the first person I interviewed for the Tesla autopilot team, which we now call the Sales AI Software Team, because it is the ad software team.

But just say safely it's a.

um,

What you say safely attend.

10 X improvement.

Improvement.

So it might almost be too much intelligence for a car I do wonder like how much intelligence should you, having a car it might get bored.

So you know it's core companies' competencies created while you wait.

So it might almost be too much intelligence for a car I do wonder like how much intelligence should you, having a car it might get bored.

Um,

and um,

Yeah actually.

You know, Optimist at scale. It is the infinite money glitch.

Actually.

And then one of the things I hope like well if we got all these caused maybe aboard.

it's like, this is a

And then one of the things I hope like well if we got all these caused that maybe our board well why they're while they're sort of if they're all board.

Or why there while there are sort of if they're onboard we could actually have a giant distributor influence fleet and say like well if they're not actively driving let's just have a giant distributed in French fleet.

A typical way to express the magnitude of... like if you've got...

um,

I actually have a giant distributed influence fleet.

something that like that.

Say like well, if they're not actively driving let's just have a giant distributed in France fleet.

You know at some point, if if you've got like tens of millions of cars in the fleet or maybe at some point 100 million cars in the fleet.

Elon Musk: It can operate tethered, so it's plugged in the whole time. That's why I call it, like, if you're true of sustainable abundance, where working will be optional. There's a limit to how much AI can do in terms of enhancing the productivity of humans. There is not really a limit to AI that is embodied. That's why I call it the infinite money glitch.

You know at some point, if if you've got like tens of millions of cars in the fleet or maybe at some point a 100 million cars in the fleet.

Elon Musk: That makes the design job radically easier and means we can delete a lot of complexity from the chip. I can't emphasize how important this is. So when you look at the various logic blocks in the chip, as you increase the number of logic blocks, you also increase the interconnections between the logic blocks. So you can think of it like there's highways. How many highways do you need to connect the various parts of the chip? And especially if you're not sure how much data is going to go between each logic block on the chip, then you kind of end up having giant highways going all over the place. It becomes an almost impossibly difficult design problem. And NVIDIA has done an amazing job of dealing with almost an impossibly difficult set of requirements. But in our case, we're going for radical simplicity.

Like, if Optimus, I think probably achieve 5x for the productivity of a person per year because it can operate at 247, um, it doesn't even need to charge. It can operate at tethered, uh, so it's it's plugged in the whole time. Um,

And.

and um,

Let's say they had at that point you know.

Let's say they had.

At that point.

Like I don't know.

So, it is that, that's why I call it. Like if you're true of sustainable abundance.

A kilowatt of inference capability of high.

um,

A kilowatt of inference capability of HIFU.

we're working with the optional.

And French capability, that's 100 gigawatts of infants.

<unk> inference capability, that's 100 gigawatts of infants.

You know, there's limited potential regarding how much AI can do in terms of enhancing the productivity of humans.

Distributed.

With power and cooling taken with with cooling and power conversion taken care of.

Um, but there is not really a limit to.

With with power and cooling taste with with cooling and power conversion taken care of.

Uh, AI, that is embodied.

So.

That's why I called it the infinite money glitch.

So.

Vaibhav Taneja: I mean, one thing which I'll further add is, people forget, like our first iteration of Autopilot was 10 years back. Elon had started this way back in the day.

That seems like a pretty significant asset.

Yeah.

Great. Thanks, and the next question comes from Adam <unk> from Morgan Stanley Adam Please feel free to yourself.

Great. Thank you on the next.

Question comes from Adam <unk> from Morgan Stanley Adam Please feel free to yourself.

I mean, one thing, which I'll further add, is that people forget our first iteration of Autopilot was 10 years back.

Elon Musk: We've got the proof to prove it.

Vaibhav Taneja: Exactly. Even on the Optimus side, as much as people think, "Okay, this is a new thing," I still remember, was it four-plus years back, we were in a finance meeting with Elon and Elon said, "Hey, our car is a robot on wheels." That's where we started developing. In fact, most of the engineering team which is working on Optimus has come from the vehicle side. That's why, you know, when we talk about manufacturing progress, we have the wherewithal because the same engineers who worked back in the day on drive units are working on actuators now. If there is any company which can do it at scale, that is going to be us.

So, you know, Elon had started this way back in the day. We got the twist to prove it.

Yeah.

And I'm going to ask your question.

Elon Musk: The net effect is that I think AI5 will be the best performance per watt, maybe by a factor of two or three, and the best performance per dollar for AI, maybe by a factor of 10. So that's, the proof's in the pudding. So obviously, we need to actually get this chip made and made at scale. That's what it looks like.

It seems like we might be having some audio issues with that and so we'll come back to you and the next question will then come from Dan <unk> from Barclays.

Exactly. And then, even on the optimistic side, right? As much as people think, "Okay, this is a new thing." I still remember, was it 4-plus years back? We were in the finance meeting with Elon, and Elon said, "Hey, our car is a robot on wheels."

Hi, Good evening, Thank you for taking the question.

And that's where we started developing. In fact, most of the engineering team that is working on Optimus has come from the vehicle side. And that's why, you know, when we talk about manufacturing progress.

I know that Tesla is really focused on with master planning for <unk>.

I know that Tesla is really focused on with master planning for bringing out instead.

We have the wherewithal because the same engineers who worked...

A world and I think we've seen over the past.

Nickel World and I think we've seen over the past.

Yeah, its willingness for Tesla to engage in and go into new markets New hands.

Yes, its willingness for Tesla to engage in and go into new markets New hands.

Travis Axelrod: Great. Thank you, Elon. We've already covered unsupervised FSD. So the next question is, "Instead of trying to replace Hardware 3 with Hardware 4, why not give an equal incentive to trade in for a new vehicle?

Elon Musk: We also have actually added a lot of new engineers as well to the team. There's actually a lot of the credit for the Optimus engineering that is also new engineers, many of them that are just out of college, actually. The Optimus engineering team is a very talented engineering team. I'd say like, wow, actually. The Optimus reviews at this point are that there's the engineering review, and then there's the manufacturing review being done simultaneously with an iterative loop between engineering design and manufacturing. We see we design something and we say like, "Oh, man, that's really difficult to make. We need to change that design to make it easier to manufacture." We've made radical improvements to the design of Optimus while increasing the functionality, but making it actually possible to manufacture. I'd say Optimus 2 is almost impossible to manufacture, frankly.

No.

When you think about the growth prospects, how do we define the areas that are really within <unk> core competency.

Versus where do you draw the line from markets or.

Versus where do you draw the line for markets or.

Lars Moravy: Yeah. We've not completely given up on Hardware 3. However, over the last year, we've offered the customers the option to transfer FSD to their new vehicle, which at times we've been running some promotions. If they got FSD, they can get better preferential rates. So we've been definitely taking care of this, but we do want to solve autonomy first, and then we'll come back with a way to take care of these customers. These customers are very important. They were the early adopters. For what it's worth, my daily commuter is a Hardware 3 car, which I use FSD on a daily basis. So we will definitely take care of you guys.

in the back in the day on drive units are working on actuators now. So that's where we can do. If there is any company, which can do it at scale that is going to be us, but we we also have actually added a lot of new Engineers as well to the team. So there's actually uh, a lot of the credit for The Optimist engineering is is actually also new new.

As many of them that are just out of college, actually, yeah.

Applications that are outside of Tesla is a core competency.

Actually I'm not sure what you mean by an AI applications outside of Tesla is a core competency.

So, uh, the Optimist engineering team is a very talented engineering team. Um,

Actually I'm not sure what you mean by an AI applications outside of Tesla is core competency.

To like, wow.

Actually. So

um,

But.

and uh,

But.

And we kind of.

We didn't have any of these core competencies when we started you know [laughter].

So it's like with zero core competencies totaling competency of zero actually.

You know, the Optimist reviews at this point are that there's the engineering review. Um,

So it's like we had zero core competencies totaling competency of zero actually.

So I mean, you can think of Tesla is like I don't know a dozen startups in one company.

And I've initiated every one of those start ups. So.

You know and and I've initiated every one of those startups so.

It's.

Travis Axelrod: Great. Thank you.

It wasn't used to make battery packs, so stationary battery packs, but now where do we make them for the homemaking for utility scale.

It wasn't used to make battery packs, so stationary battery packs, but now where do we make them for the homemaking poor utility scale.

Ashok Elluswamy: addition, once the V14 release series is fully done, we are planning on working on a V14 Lite version for Hardware 3, probably expected in Q2 next year.

Oh, well Mega pack.

Uh huh.

We've created the supercharge network globally.

<unk>.

We've created the supercharge network globally.

Elon Musk: By my two bypass point, we've gone from a person in a robot outfit to what people have seen with Optimus 2.5 where it's doing Kung Fu. Optimus was at the Tron premiere doing Kung Fu, just out in the open, like with Jared Leto. Nobody was controlling it. It was just doing Kung Fu with Jared Leto at the Tron premiere. You can see the videos online. The funny thing is a lot of people walked past it thinking it was just a person. Even though with Optimus 2.5, you can see that it has a waist that's three inches wide, which is obviously not a human. The movements were so human-like that people didn't realize, a lot of people didn't realize they were looking at a robot. What I'm saying is Optimus 3 will be a giant improvement on that and made at scale.

And then there's the manufacturing review, um, being done. It's, it's simultaneously, um, with an iterative Loop between engineering design and and Manufacturing because then we see we, we we designed something, and we see like, oh man, that's really difficult to make. We need to change that design, to make it easier to manufacture. Um, so we've made radical improvements to the design of Optimus, um, while increasing the functionality but making it actually possible to manufacture. Like I'd say, Optimus 2 is almost impossible to manufacture frankly. Um,

I don't want no one else has created a global Supercharge network in fact that North American supercharger network. So good at that that basically that.

but um,

I don't want no one else has created a global supercharge network in fact that North American so you've talked to that were so good at that that basically but.

Travis Axelrod: Awesome. Thanks, Ashok. All righty. Our final question from Sei is, "How long until we see self-driving Tesla semi-trucks? And could you see this technology replacing trains?

My $2 million point. We've gone from...

Uh, you know, a person in a robot output to...

Every other manufacturer in North America is converted to our standard.

Yeah every other manufacturer in North America is converted to our standard.

And use our that the Tesla supercharger network.

And us are that the Tesla supercharger network.

Elon Musk: Yeah. So I guess I'll start with that in terms of the semi production plan and schedule. So the factory is going on schedule. We've completed the building and are installing the equipment now. We've got our fleet of validation trucks driving on the road. We'll have larger builds towards the end of this year and then our first online builds in the first part of next year, ramping into the Q2 timing with real volume coming in the back half of the year. So that's going quite well, and that's the first step, obviously, getting autonomous trucks on the road. In terms of the trains, they're really great for long point-to-point deliveries. They're super efficient, but that last mile, the load unload, can be better served for shorter distances with autonomous semis, and that would be great.

Yeah. So I guess I'll start with that in terms of the semi production plan and schedule. So the factory is going on schedule. We've completed the building and are installing the equipment now. We've got our fleet of validation trucks driving on the road. We'll have larger builds towards the end of this year and then our first online builds in the first part of next year, ramping into the Q2 timing with real volume coming in the back half of the year. So that's going quite well, and that's the first step, obviously, getting autonomous trucks on the road. In terms of the trains, they're really great for long point-to-point deliveries. They're super efficient, but that last mile, the load unload, can be better served for shorter distances with autonomous semis, and that would be great.

But if it was so easy why don't they just do it.

Uh, what, what people have seen with Optimus 2.5, where it's doing Kung Fu? Um, you know, it was like Optimus was at the, uh, at the Toronto Premiere. Um,

And the chip design team.

Doing Kung Fu, you know, just out in the open, you know, like with Jared Leto.

We started that from scratch the Tesla AI software team with Sutter from scratch.

We started that from scratch the Tesla AI software team was set up from scratch.

like there wasn't, uh,

I literally just saying hey, we're going to start this thing I posted on Twitter now X and then.

Nobody was controlling it. It was just doing Kung Fu with Jared Leto. Uh, and, you know, at the Crown premiere.

Um, you can see the videos online.

um,

Join us if you'd like to build it.

Yeah join us if you'd like to build it.

In fact, the choke was I believe the first person I interviewed for the Tesla autopilot team, which we're now called Seal's AI software team.

In fact, the shock was I believe the first person I interviewed for the Tesla autopilot team, which we're now called says AI software team.

And, um, actually, the funny thing is, like a lot of people walked past it, thinking it was just a person.

even though with Optimus,

We have to ask soccer team.

As soccer team.

Uh, 2.5. You can see that it has a, you know, a waist that's 3 inches wide, which results in not a human.

So you know its core companies' competencies created while you wait.

um, so

And.

Uh, but but the movements were so humanlike, that people didn't realize, a lot of people didn't realize they were looking at a robot.

Elon Musk: And so we do expect that to probably shift as we really, as Elon said, change the way transportation is considered. And so we're looking forward to that timeline. And Ashok, I know you can take the Full Self-Driving part.

And so we do expect that to probably shift as we really, as Elon said, change the way transportation is considered. And so we're looking forward to that timeline. And Ashok, I know you can take the Full Self-Driving part.

so,

[laughter] Optimus that scale is the infinite money glitch.

um,

Optimists at scale is the infinite money glitch.

It's like this is.

And what I'm saying is, like, Optimus 3 will be...

It's like this is.

a giant Improvement on that.

It's difficult to express the magnitude of like if you bought.

It's difficult to.

Elon Musk: Like I said, a very difficult thing. The Optimus sort of engineering and manufacturing reviews, and there's the Friday night meeting with Optimus, which sometimes goes till midnight. My Saturday meeting is with the Saturday afternoon is with the AI-5 chip design team. Those two things are crucial to the future of the company.

Um, and made it scale.

Express the magnitude of like if you've got.

Ashok Elluswamy: Currently, the team is super focused on solving for passenger vehicles autonomy. That said, the same technology will extend quite easily to the semi-truck once we have a little bit of data from the semi-trucks.

Something like that.

A very difficult thing. Um,

Something like that.

Pat.

Like if a optimists I think probably cheap five X the productivity of a person per year because it can operate.

yeah, the The Optimist, uh,

Like if autonomous I think probably cheap IBEX the productivity of a person per year because it can operate 20.

24 seven.

It doesn't even need to charge it can operate a tethered.

Sort of injuring and manufacturing reviews. And there's the Friday night meeting with Optimus, which sometimes goes to midnight.

It doesn't even need to charge it can operate a tethered so it's.

um, and then

Travis Axelrod: Great. And now we will move over to analyst questions. The first question comes from Emmanuel at Wolfe. Emmanuel, please go ahead and unmute yourself.

It's plugged in the whole time.

<unk>.

And.

My Saturday meeting is what that is, is The Saturdays with the AI 5 Tripp design team.

What is it that's that's why I call. It like if you true sustainable abundance.

Um,

What is it that's that's why I call. It like if you drove sustainable abundance.

[Analyst] (Wolfe Research): Great. Thanks so much. Hi, everybody. So Elon, you talked about expanding production of vehicles as fast as possible now that you have confidence in the unsupervised autonomy. How should we think about that in the context of your existing capacity of 3 million units? Is that where you're hoping to get volume to? What sort of timeline are we talking about? And would this require some level of boosting or incentivizing demand? Would this basically be prioritizing volume over near-term profitability given the longer-term opportunity?

So, um, those two things are.

We're working will be optional.

Uh, crucial to the future of the company.

You know that there's limits to how much how much I can do in terms of enhancing the productivity of humans.

Travis Axelrod: Great. Dan, did you have a follow-up?

That there's limits to how much you can do in terms of enhancing the productivity of humans.

Dan Levy: Yeah, I think just as a related, maybe you could just talk about to what extent are the AI efforts at Tesla and XAI complementary, or are they just different forms of AI. Maybe you can just help distinguish for the audience. Thank you.

But there is not really a limit to.

Great. And Dan, did you have a follow-up? Yeah, I think just as a related question.

But there is not really a limit to.

AI that is embodied.

That is embodied.

That's why I called infinite money glitch.

That's why I called in for money quit.

Elon Musk: Yeah. There are different forms of AI. The xAI, so Grok is like a giant model that you could not possibly squeeze Grok onto a car. That's for sure. It is a giant beast of a model. With Grok, it's trying to solve for artificial general intelligence with a massive amount of AI training, compute, and inference compute. For example, Grok 5 will actually only run effectively on a GV300. That's how much of a beast that Grok 5 is. Tesla's models are, I don't know, maybe about less than 10% the size, maybe closer to 5% the size of Grok. They're really coming at the problem from very different angles. xAI and Grok, you know, they're competing with Google Gemini and OpenAI ChatGPT and that kind of thing. Some of it's complementary. For example, for Grok voice, being able to interact with Grok in the car is cool.

I mean, one thing, which I'll further that is I mean people forget like our philosophy, Jason of autopilot was 10 years back. So you know you had started this way back in the door.

I mean, one thing, which I'll further that is.

Maybe you could just talk about to what extent the AI efforts at Tesla and X.A.I. are complementary, or if they are just different forms of AI? Maybe you can just help distinguish where the audience. Thank you. Yeah, they are different forms of AI. Um,

Forget like our philosophy, Jason of autopilot was 10 years back.

so,

The.

you know, the

You know you had started this way back and then you've got the trees to prove it.

You've got the trees to prove them exactly.

Exactly and then even even on the Optima side.

And then even even on the optimum side right as much as people think okay. This is a new thing I still remember what was it four plus years back we went into finance meeting with Atlanta in Atlanta said, hey cause the robot on wheels.

As much as people think okay. This is a new thing I still remember was it four plus years back we went into finance meeting with Atlanta and he loves that.

Elon Musk: Well, our capacity isn't quite three million, but it will be three million at some point. Aspirationally, it could be three million within, we could probably hit an annualized rate of three million within 24 months, I think, maybe less than 24 months. Bearing in mind there's an entire supply chain, a vast supply chain that's got to also move in tandem with that. So we're going to expand production as fast as we can and as fast as our suppliers can sort of keep up with it. And then we're going to think about where do we build incremental factories beyond that. The single biggest expansion in production will be the Cybercab, which starts production in Q2 next year. That's really a vehicle that's optimized for full autonomy.

Cause the robot on wheels.

And that's where we started developing in fact, most of the engineering team, which is working on optimism.

The actually I sort of grock is like a a giant model that that uh, you could not, you could not possibly squeeze grok, uh, onto a car because that's for sure. It is a giant Beast of a model. It's with, with grok. It's trying to say solve for artificial general intelligence with a massive amount of AI training compute, and, and inference compute.

um,

Ask him from the weak outside and that's why when we talk about manufacturing progress.

Ask him from the weak outside and that's why you know when we talk about manufacturing progress.

So, for example, for Grow 5, it will actually only run effectively on a GV30.

We have the wherewithal because the same engineers who worked.

That's how much of a boost that Grow 5 is, uh,

Back in the day on drive units are walking on actuators now so that's where we can do if there is any company, which can do it at scale that is gonna be us, but we also have actually added a lot of new engineers as well the team. So there's actually a lot of the credit.

um, so, uh

Back in the day on drive units are working on actually it does now so that's where we can do if there is any company, which can do it at scale that is gonna be us, but we also have actually added a lot of new engineers as well as the team. So there's actually a lot of the credit.

No worries Tesla's. Uh,

You know, models are, I don't know, maybe about less than 10% the size, maybe closer to 5% the size of grog.

um,

so,

For the Artemis Engineering is actually also near you introduced many of them that are just out of college actually.

For the Optimus Engineering is actually also near you introduced many of them that are just out of college actually yeah.

Yeah, they're really coming at the problem from very different angles.

um,

So.

So a doctor with engineering.

Dr. Louis Engineering team.

Xan, Grark are, you know, they're competing with, um,

Team is a very talented team.

Team is a very talented engineering team.

uh,

I'd say like Wow.

Actually so.

Your Google Gemini and opening, add ChatGPT and that kind of thing. Um,

And.

The Artemis reviews at.

Elon Musk: It in fact does not have a steering wheel or pedals and is really an engineering optimization on minimizing cost per mile for fully considered cost per mile of operation. So for our other vehicles, they still have a little bit of the horseless carriage thing going on where obviously if you've got steering wheels and pedals and you're designing a car that people might want to go very fast acceleration and tight cornering, like high-performance cars, then you're going to design a different car than one that is optimized for a comfortable ride but doesn't expect to go past sort of 85 or 90mi/h and is just aiming for a gentle ride the whole time. That's what Cybercab is. So yeah, so do I think we'll sacrifice margins? I don't think so. I think the demand will be pretty nutty.

You know the Artemis reviews.

At this point or does the engineering review.

At this point are there.

The Engineering review.

Elon Musk: Grok for Optimus voice recognition and voice generation is Grok. That's helpful there. They are coming at it from kind of opposite ends of the spectrum.

So um and some of its complimentary. I mean, if we could, for example, for grok voice, uh be able to interact with grok in the car is cool. Um,

And then there's the manufacturing reviewed a mcdonalds.

Simultaneously.

With an iterative loop between engineering design and manufacturing because when we see it.

We design something that you feel like a man that's really difficult to make we need to change that design to make it easier to manufacture.

We designed something here, we say like Oh man, that's really difficult to make we need to change that design to make it easier to manufacture.

Brock, uh, for, you know, Optimus voice recognition and what voice generation is, Rock. So, that's helpful there. But they are, uh, coming at it from kind of opposite ends of the spectrum.

Travis Axelrod: All righty. Adam, let's give it another try. When you're ready, please unmute yourself for the next question. All righty. Unfortunately, still having audio issues. We're going to move on to Walt from Lightshed. Walt, please go ahead and unmute yourself.

So we've made radical improvements to the design of Optimus.

So we've made radical improvements to the design of Optimus while.

Already. Um, add on, let's give it another try. Uh, when you're ready, please unmute yourself for the next question.

While increasing the functionality, but making it actually possible to manufacture like I'd say optimists too is.

While increasing the functionality, but making it actually possible to manufacture like I'd say optimists to us.

Almost impossible to manufacturer frankly.

But.

<unk>.

But.

All righty, unfortunately, uh, still having audio issues. So we're going to move on to, uh, Walt from Lightshed.

Two five point, we've gone from.

A two five point we've gone from.

A person in a robot out but to.

What people have seen with office 2.5, where it's doing Kung Fu.

Walt, please go ahead and unmute yourself.

What what people have seen with utmost 2.5, where it's doing Kung Fu.

[Analyst 2]: Can you hear me now?

Travis Axelrod: Yes.

No it was like.

Can you hear me now?

[Analyst 2]: Perfect. Thank you. Just getting back to Austin, if you can remove the safety driver at your end, is the limitation in the Bay Area just regulatory, or is it kind of the market-by-market learning process? I guess similarly, in the 8 to 10 markets that you mentioned to get added, is the decision there to put a safety attendant in the passenger seat or the safety driver in, is that like your step-by-step process to opening up a market, or is it really just the regulation in the individual market?

I always like Dr.

This was at the at the Trop and Premier.

After this was at the at the Trop and Premier doing Kung Fu.

Doing Kung Fu.

Up in the open you know like with Jared Leto [laughter] like there wasn't a.

And the open you know like with Jared Leto [laughter] like there wasn't a.

Elon Musk: Here's the killer app, really. What it comes down to is, can you text while you're in the car? And if you tell someone, "Yes, the car is now so good," you can be on your phone and text the entire time while you're in the car. Anyone who can buy the car will buy the car and a story. So that's what everybody wants to do. In fact, not everyone wants to do it. They do do that. And that's why, in fact, the reason you've seen there's been an uptick in accidents pretty much worldwide is because people are texting and driving. So autopilot actually dramatically improves the safety here because if somebody's looking down at their phone, they're not driving very well. So that's really the game changer.

Nobody was controlling it it was just doing Kung Fu hero plateau.

But nobody who is controlling it it was just doing Kung Fu hero plateau.

At the <unk> Premier.

You know at the Crown Premier.

You can see the videos online.

And actually the only things like a lot of people walked past it.

Yes, so perfect. Thank you. Um, um, just getting back to Austin if you can remove the safety driver at your end. Um, is the limitation the Bay Area just regulatory or is it kind of the market by market learning process and I guess similarly in the 8 to 10 markets that you mentioned to get added, is the decision there to put, you know, a safety attendant in the passenger seat.

And actually the funny thing is like a lot of people walked past it.

Or the safety driver in. Is that?

Thinking it was a.

Thinking it was a just.

Just a person.

Even though with documents.

Even though with darkness.

Like your step-by-step process to opening up a market? Or is it really just the regulation in the individual market?

Elon Musk: I think even if the regulators weren't making us do it, we'd still do that as the right sort of cautious approach to a new market. Just to make sure that we're being paranoid about safety, I think it makes sense to have either a safety driver or safety occupant in the car when we first go to new markets to confirm that there's not something we're missing. All it takes is like 1 in 10,000 trips to go wrong, and you've got an issue. It's just to make sure, like is there some peculiarity about a city, like a very difficult intersection or, I don't know, something that's an unexpected challenge in a city for that 1 in 10,000 situation? I think we probably could just let it loose in these cities, but we just don't want to take a chance.

At 245, you can see that it has a wasted three inches wide, which was obviously not a human.

So.

But the movements were so human like that people didn't realize a lot of people didn't realize they were looking at a robot.

But the movements were so human like that people didn't realize a lot of people didn't realize they are looking at a robot.

Well, it's I think I think even if The Regulators weren't making us do, it, it would still do that as the as the sort of right? Sort of caution cautious approach to a new market. So just to

So.

And what I'm, saying is like Artemis through will be at.

And what I'm, saying is like Artemis through will be.

Make sure that we're being, you know, paranoid about safety. I think it makes sense to have a sort of, uh,

Giant improvement on that.

And made it scale.

But like I said, it's a very difficult thing.

A sort of either safety driver or safety occupant in the car. When we first go to new markets, we just want to confirm that there's not something we're missing.

Yeah the Optimus.

Yeah the optimists.

<unk>.

Elon Musk: We do see at this point, I feel essentially 100% confident, I say not essentially, 100% confident, that we can solve unsupervised Full Self-Driving at a safety level much greater than human. So we've released 14.1. We've got a technology roadmap that's, I think, pretty amazing. We'll be adding reasoning to the car. Our world simulator for reinforcement learning is pretty incredible. When you see the Tesla reality simulator, you can't tell the difference between the video that's generated by the Tesla reality simulator and the actual video. It looks exactly the same. So that allows us to have a very powerful reinforcement learning loop to further improve the Tesla AI. We're going to be increasing the parameter count by an order of magnitude. That's not in 14.1. There are also a number of other improvements to the AI just that are quite radical.

Sort of engineering and manufacturing reviews, and that's what that's.

Sort of engineering and manufacturing reviews, and that's what that's the Friday night meeting with Optimists, which sometimes goes till midnight.

Um, because all it takes is like 1 in 10,000 trips to go wrong, and you've got an issue.

so, um,

The Friday night meeting with Optimists, which sometimes goes till midnight.

And then.

Saturday meetings is what that is a Saturday afternoon with the the AI five chip design team.

Saturday meetings is what is the Saturday afternoons with the the AI five chip design team.

It's just to make sure, like, is there some curiosity about a city, like a very difficult intersection? Um, or, I don't know, something that's an unexpected challenge?

So.

Uh, in a city. Uh, for that 1 in 10,000 situation. Um,

Those two things are.

so,

Those two things are.

um,

Crucial to the future of the company.

Elon Musk: What we're talking about here is maybe three months of safety driver in a new metro to confirm that it's good, and then we take the safety driver out, that kind of thing.

Great Danes, you'll follow up yeah.

Green dangers you'll follow up yeah.

Related.

Maybe you could just talk about to what extent are the AI efforts at Tesla and Xa add complementary or are they just different forms maybe you could just help distinguish the audience. Thank you, yes that there are different forms of AI.

Maybe you could just talk about to what extent are the AI efforts at Tesla and Xa add complementary or are they just different forms maybe you could just help distinguish the audience that.

[Analyst 2]: Okay. On Full Self-Driving version 14, it has a different feel than 13, and it's also, I think, a little different than what it feels like in Austin. Is it basically a different development path that you're doing in terms of the robotaxi stuff versus what you're dropping to the early adopters? When you push these new builds, is it that you're looking for notable improvements in intervention rates, or is that largely solved and it's more about adding the functionality, like the parking, the drive mode, or just the overall comfort?

Take the safety driver off that, that kind of thing.

There are different forms of AI.

So.

The.

Yes.

Yeah.

They actually had a.

They actually had a crock is like a giant model that that you cannot you cannot possibly squeeze croquis onto a car that's for sure. It is a giant piece of a model. It's with Crockett is trying to say so for artificial general intelligence with a massive.

Croc is like a giant model that that you cannot you cannot possibly squeeze.

Okay. And then on, on FSD 14, it has a different feel than 13. And it's also I think a little different than what it feels like in in Austin, are you, is it basically different development, paths path that you're doing in terms of the robo, taxi stuff versus what you're dropping to the early adopters and when you, and when you push these new bills,

Onto a car that's for sure. It is a giant piece of a model. It's with Crockett is trying to say so for artificial general intelligence with a massive amount of AI training compute and inference compute.

Elon Musk: The first priority when we release a major new software architecture for Autopilot is safety. It starts off with, you know, obviously, safety prioritized, and then we solve comfort thereafter, which is why I don't recommend people take the initial version. That's why I say, like, you know, most people should wait until 14.2 before they actually download version 14. Because by 14.2, we will have addressed many of the comfort issues. The priority is very much safety first, and then thereafter the comfort issues. That's why most people are like, "I'd probably. It'll be a little like it'll be safe, but jerky." We just need time to kind of smooth the rough edges and solve for comfort in addition to safety with a major new Autopilot architecture change.

Is it that you're looking for notable improvements in intervention rates, or is that largely solved? And it's more about adding the functionality like the parking and the drive modes, or just the overall comfort?

The amount of AI training compute and inference compute.

So for example for rock five will actually only run effectively on at GBP 300.

So for example for rock five will actually only run effectively on a GBP 300.

Elon Musk: So this car will feel like it is a living creature. That's how good the AI will get with the AI4 computer before AI5. And then AI5, like I said, is by some metrics 40 times better. Let's just say safely, it's a 10x improvement. So it might almost be too much intelligence for a car. I do wonder how much intelligence should you have in a car? It might get bored, actually.

How much of a beast that <unk> five has.

So.

And at worst Teslas I.

Whereas Tesla has.

Hi.

Models are I don't know maybe about.

Yeah.

No, the first priority when we release a, a major new software architecture for autopilot is safety. So, so it's it starts off with, with safety, you know, obviously, safety prioritized, and then we've and then we solve Comfort thereafter. Which is why I don't recommend people take the the initial version, like, like,

Or I don't know maybe about.

Less than 10% the size maybe closer to.

5% the size of Av block.

5% the size of clock.

So yeah.

So.

Yeah, they're they're they're they're really coming at the problem from very different.

Yeah, they're they're they're they're really coming at the problem from very different angles.

That's why I say, like, you know, most people should wait until 4:42 before they actually download version 14. By 14.2, we will have addressed many of the comfort issues.

Angles.

X and Croc are there competing with them.

Elon Musk: Then one of the things I thought of, well, if we've got all these cars that maybe are bored, well, while they're sort of, if they are bored, we could actually have a giant distributed inference fleet and say, "Well, if they're not actively driving, let's just have a giant distributed inference fleet." At some point, if you've got tens of millions of cars in the fleet, or maybe at some point, 100 million cars in the fleet, and let's say they had at that point, I don't know, a kilowatt of inference capability, high-performance inference capability, that's 100 gigawatts of inference distributed with power and cooling taken care of. So that seems like a pretty significant asset.

Exxon Crocker or they're competing with them.

Google Gemini and opening a chat GBT and that kind of thing.

Um, the priority is very much safety first and then thereafter the comfort issues. That's why most people are like, "Yeah, I probably... it'll be a little, like, it'll be safe but jerky."

Can you Cook with Gemini and opening a chat GBT and that kind of thing.

So.

Some of it's complementary for example for Crock voice.

Some of it's complementary that you would be pretty good for example for Crock voice.

Um, and, uh, we just need time to kind of smooth the rough edges. Um, and sulfur compound, in addition to safety, with a major news, uh,

Mailed interact with Kraken the cars cool.

Interact with Kraken the cars cool.

Autopilot architecture, uh, change.

Elon Musk: It really is, I mean, I know what the, you know, the roadmap is for the Tesla real-world AI and at a very granular detail. Obviously, Ashok Elluswamy is leading that. I mean, I spend a lot of time with the team going, you know, in excruciating detail here on what we're doing to improve the real-world AI. Like I said, this car is going to feel like it is a living creature. That's with AI-4 before even AI-5.

um,

Uh huh.

For Optimus voice.

<unk> voice.

But, uh, it really is, uh,

Voice recognition and.

What are your voice generation.

I mean, I know what the...

What are your voice generation.

It's rock so that's that's helpful. There.

As crocs. So that's that's helpful. There, but they are are coming at it from kind of opposite ends of the spectrum.

You know, the roadmap is for the Tesla real-world AI, and

They are are coming at it from kind of opposite ends of the spectrum.

Yeah.

And a brand new granular data. Detail, obviously a show is leading that, um,

Yeah.

Alrighty, Adam let's give it another try.

And I, and I, I mean, I spent a lot of time with the team going, you know, in.

When you are ready please unmeet yourself for the next question.

like,

Excruciating detail here on and on.

Alrighty, Unfortunately, as still having audio issues. So we're going to move on to our Walt from light shed.

What? What? We're doing to improve the real-world AI.

Alrighty, Unfortunately, as still having audio issues. So we're going to move on to our Walt from light shed.

Um,

And, um, like I said, this car is going to feel like it is a living creature.

Ashok Elluswamy: Yeah. The roadmap is super exhilarating. It's like we're waiting so much to release all the stuff we are working on. In terms of what we ship to customers versus robotaxi, it's mostly the same. Obviously, customers have somewhat features like, you know, they can choose whether the car wants to park in a spot or in a driveway or something like that, which is not super relevant for robotaxi. There's only a few minor changes like those ones. The majority of the algorithms and architecture, everything is the same between those two platforms.

And that's what the AI is for, before, even AI 5.

Please go ahead and them yourself.

Travis Axelrod: Great. Thanks, Elon. The next question comes from Adam from Morgan Stanley. Adam, please feel free to unmute yourself. Adam, go ahead and ask your question. Seems like we might be having some audio issues with Adam. So we'll come back to you. The next question will then come from Dan from Barclays.

Please go ahead and them yourself.

Can you hear me now.

Yes, perfect. Thank you.

Yes.

Just getting back to Austin, if you can remove the safety driver at year end is the limitation in the Bay area, just regulatory or is it kind of a market by market learning process and I guess similarly in the eight to 10 markets that you mentioned to get at it is the decision there to put a safety attended in the passenger seat.

Elon Musk: Yeah, as I mentioned earlier, we'll be adding reasoning to, I don't know, Ashok, is that like reasoning in 14.3, maybe 14.4, something like that?

Yeah, the door Drive is super exhilarating like it's like so like waiting so much, like, at least all the stuff we are working on in terms of like, you know what, we ship to customers versus robot taxi, uh, it's more mostly the same. Obviously, customers have some more features like, you know, they can choose the car wants to park in a spot or drive where something like that, which is not super relevant for robot taxi. Uh, but there's only like few, minor changes like those ones, but the majority of the algorithms and architecture, everything is the same between those 2 platforms.

Or the safety driver and is that.

Daniel Roeska: Hi. Good evening. Thank you for taking the question. Elon, I know that Tesla is really focused on, with Master Plan 4, bringing AI into the physical world. And I think we've seen over the past, this willingness for Tesla to engage and go into new markets, new TAMs. So when you think about the growth prospects, how do we define the areas that are really within Tesla's core competency versus where do you draw the line for markets or AI applications that are outside of Tesla's core competency?

Like your step by step process to opening up a market or is it really just the regulation in the individual market.

Your step by step process to opening up a market or is it really just a regulation in the individual market.

Ashok Elluswamy: Yeah, by the end of this year, for sure.

Elon Musk: Yeah. With reasoning, it's literally going to think about which parking spot to pick. It's going to say, "This is the entrance, but actually, probably there's not a parking spot right at the entrance if it's full, you know, if the parking lot is fairly full, the probability of an open parking spot right at the entrance is very low." What it'll simply do is drop you off at the entrance of the store and then go find a parking spot. It's going to get very smart about figuring out a parking spot. It's going to spot empty spots much better than a human because it's got 360-degree vision. Like I said, it's going to use reasoning to solve things.

Well I think I think even if the regulators weren't making us do it we'd still do that as the sort of right sort of course cautious approach to a new market. So just.

Well I think I think even if the regulators weren't making us do it we'd still do that as the S. B sort of right sort of course cautious approach to a new market. So just.

Yeah. But, you know, as I mentioned earlier, we'll be adding reasoning too. Um, I don't know. Sure. Is that like reasoning in 14.3? Maybe 14.4, something like that. Yeah, by another year, for sure. Yeah.

Make sure that we're being paranoid about safety.

I think it makes sense to have it sort of.

We think it makes sense to have it sort of.

I sort of either safety driver safety occupants in the car.

When we first go to new markets just to confirm that that's not something we're missing.

When we first go to new markets just to confirm that theres not something we're missing.

So with reasoning. It's literally going to think about which parking spot to pick, uh, at the. So it's going to say this is the entrance but actually probably there's not a parking spot right at the entrance. If it's a, a full, you know, if the if the parking lot is fairly full, the problem of open parking spot right at the entrance is very low. Um, but actually, what I'll simply do is drop you off at the entrance of the store and then go find a parking spot.

Um,

Because all it takes is like one in 10000 trips to go wrong and you got it you got an issue so.

Elon Musk: Actually, I'm not sure what you mean by AI applications outside of Tesla's core competency. But we didn't have any of these core competencies when we started. So it's like we had zero core competencies, total competency of zero, actually. So I mean, you can think of Tesla as, I don't know, a dozen startups in one company. And I've initiated every one of those startups. So we're used to make battery packs, stationary battery packs, but now we do. We make them for the home, make them for utility scale with Powerwall, Megapack. We created the Supercharger network globally. No one else has created a global Supercharger network. In fact, our North American Supercharger network is so good at that basically that every other manufacturer in North America has converted to our standard and uses the Tesla Supercharger network.

It's it just to make sure that it is there some.

It's it just to make sure that it is there some kill you already about a city like a very difficult intersection.

but it's it's going to get very smart about figuring out a parking spot. I just going to spot figure out. It's going to spot empty spots, much better than the human. It's got 360 degree Vision. Um,

Or you're already about a city like a very difficult intersection or I don't know something that's that's an unexpected challenge.

And it's going to. Yeah, like I said, just.

Or I don't know something that's that's an unexpected challenge.

Ashok Elluswamy: Fitting that all inside the computer that has the AI-5 chip is the actual challenge. That's what the team is working on. Obviously, you can do reasoning on the server that takes forever, but then in the car, you need to make real-time decisions. Fitting all that into the computer that's in the car, that's the challenge.

It's going to re use reasoning to solve things.

In a city put out one in 10000 situation.

And the city put out one in 10000 situation.

So.

I think we'd probably if you could just let it loose in this in these studies, but we just don't want we don't want to take a chance and and like you know what we.

I think we probably could you just let it loose in this in these studies, but we just don't want we don't want to take a chance and alike.

Elon Musk: Yeah, that's why I say I have a pretty good understanding of like AI, you know, the sort of the giant model level with Grok and with Tesla. I'm confident in saying that Tesla AI has the highest intelligence density. When you look at the intelligence per gigabyte, I think Tesla AI is probably an order of magnitude better than anyone else. It doesn't have any choice because that AI has got to fit in the AI for computer. The discipline of having that level of AI intelligence density will pay great dividends when you go to something that has an order of magnitude more capability like AI-5. Now you have that same intelligence density, but you've got 10 times more capability in the computer.

And putting that all inside the computer that has the A4 is the actual challenge. That's what the team is working on because, obviously, you can do reasoning on the server that takes whatever, but then in the car, you need to make real-time decisions. So, putting all that into the computer that's in the car, that's the challenge.

We're talking about here is.

Talking about here is.

Maybe three months of safety driver in a new metro ticking from that it's good and then I would take the safety driver of that that kind of thing.

Maybe three months of safety driver in a new metro taking from that it's good and then I would take the safety drive Ralph that that kind of thing.

Okay, and then on an FSD 14, it has a different field in 13, and it's also I think a little different than what it feels like and in Austin.

Yeah that's why I say like like I have a pretty good understanding of like AI, you know, the sort of the giant model level with grock and with with Tesla and like I'm confident in saying that Tesla has the the Tesla AI has the highest intelligence density. When you look at the, the intelligence per gigabyte,

um,

Is it basically different development paths path that youre doing in terms of the robo taxi stuff versus what youre dropping to the early adopters and when you and when you push these new builds.

I think, like Tesla, AI is probably an order of magnitude better than anyone else. Um, and it doesn't have any choice because that AI has got to fit in the AI for computer.

Um,

the, but the

Elon Musk: But if it was so easy, why don't they just do it? The chip design team started that from scratch. The Tesla AI software team started from scratch. I literally just say, "Hey, we're going to start this thing." I post it on Twitter, now X, and then, "Join us if you'd like to build it." In fact, Ashok was, I believe, the first person I interviewed for the Tesla Autopilot team, which we now call the Tesla AI software team, because it is the AI software team. So it's core competencies created while you wait. And Optimus at scale is the infinite money glitch. It's difficult to express the magnitude of if you've got something that if Optimus, I think, could probably achieve 5x the productivity of a person per year because it can operate 24/7. It doesn't even need to charge. It can operate tethered.

Is it that you're looking for notable improvements in intervention rates or is that largely solved and it's more about adding the functionality like the parking to drive knowledge or just the overall comfort.

The discipline of having that level of AI.

Intelligence density will pay great dividends when you go to something that has an order of magnitude more capability, like AI 5.

The first priority when we released.

The first priority when we release.

now, you have that same intelligence density, but

A major new software architecture for autopilot is safety. So so it starts off with safety, obviously safety prioritized and then we and then we saw comfort thereafter, which is why I don't recommend people take the initial version.

Major new software architecture for autopilot is safety. So so it starts off with safety, obviously safety prioritized and then we and then we solve comfort thereafter, which is why I don't recommend people take the initial version.

But you got 10 times more capability in the computer.

Travis Axelrod: Great. The next question will come from Colin at Oppenheimer. Colin, please unmute yourself when you're ready. Colin, go ahead and unmute yourself, please.

From Colin at Oppenheimer. Uh, Colin, please unmute yourself when you're ready.

Lars Moravy: Thanks so much, guys. I appreciate you bringing up the challenges of hand dexterity and humanoids, along with the complexity of the supply chain and the vertical integration you guys are pursuing. I'm just trying to harmonize the timeline for the start of production next year with the current state of the supply chain and what sounds like a fair amount of work remaining on the dexterity before you can really freeze the hardware design and start to scale up production.

<unk>.

Like.

That's why I say like most people should wait until for 14.24 before they actually download version 14.

That's why I say like most people should wait until for 14.24 before they actually download version 14, because by 14.2, we will have addressed many of the comfort issues.

Because by 14.2 will address many of the comfort issues.

Hmm.

Priority is very much safety first and then thereafter.

Hmm.

Priority is very much safety first and then thereafter.

The comfort issue, that's why most people Oh, my God, I, probably that'll be a little like it'll be safe, but jerky.

I'll go ahead and unmute yourself. Please thanks so much guys. Um, you know, I appreciate you bringing up the the challenges of hand dexterity and humanoids, um, you know, along with the complexity of the supply chain and the, the vertical integration and you guys are pursuing, you know, I'm just trying to harmonize the the timeline for the start of production, you know, next year with the current state of the supply chain and what it sounds like a fair amount of work remaining on the dexterity before. You can really freeze the hardware design and and start to scale up production.

And I.

And I would just need time to kind of smooth the rough edges.

Elon Musk: The hardware design will not actually be frozen even through startup production. There will be continued iteration because a bunch of the things that you discover are very difficult to make. You only find that pretty late in the game. We will be doing rolling changes for the Optimus design even after startup production. I do think that the new hand is an incredible piece of engineering. Actually, we will have a production intent prototype ready to show off in Q1, probably February or March. We are going to be building a million-unit Optimus production line, hopefully with the production start towards the end of next year. That production ramp will take a while to get to an annualized rate of a million because it is going to move as fast as the slowest, dumbest, least lucky thing out of 10,000 unique items. It will get to a million units.

We just need time to kind of smooth the rough edges.

And so if a company in addition to safety with the with the major news.

And sulfur comfort in addition to safety with a with a major news.

Um, well, the hardware design will not actually be frozen.

Yeah, you know what.

The pilot architecture change.

The autopilot architecture change.

But it really is.

Elon Musk: So it's plugged in the whole time. And so that's why I call it if you're true of sustainable abundance, where working will be optional. There's a limit to how much AI can do in terms of enhancing the productivity of humans. But there is not really a limit to AI that is embodied. That's why I call it the infinite money glitch.

But it really is.

I mean, I know what the you know.

I mean, I know what the.

The roadmap is for the test the real World AI and <unk>.

The roadmap is for the test the real World AI and.

And that's pretty granular detail, obviously, a show cause leading that.

And at a pretty granular detail, obviously, a show cause leading that.

And I I mean, I spent a lot of time with the team.

Even through startup production, um, there will be continued iteration, um, because a bunch of the things that you discover are very difficult to make. You only find that pretty late in the game. So we'll be doing rolling changes for The Optimist design, even after solid production. Um, but I do think that the, you know, the new hand is...

And I I mean, I spent a lot of time with the team going.

Going in.

Um, an incredible piece of engineering.

Like Dude.

<unk>.

and uh,

Like.

Excruciating detail here.

<unk>.

What we're doing to improve the real world AI.

You know? That's, yeah. Well, I say we'll have, um, a production intent prototype.

And I'm like I said, it's this car is going to feel like it is a living creature.

And I'm like I said, it's this car is going to feel like it is a living creature and.

Ready to show off in, you know, Q1, probably February or March. Um,

That's what AI for before even a high five.

[Analyst] (Wolfe Research): I mean, one thing which I'll further add is, I mean, people forget our first iteration of Autopilot was 10 years back. So Elon had started this way back in the day.

Vaibhav Taneja: I mean, one thing which I'll further add is, I mean, people forget our first iteration of Autopilot was 10 years back. So Elon had started this way back in the day.

And that's what the AI for it before even a high five.

and then we're, uh,

Yes.

Yeah that orbcomm has so far anything or anything like that.

Or anything or anything like that.

Yeah, we're going to be building a million units.

Hello.

So much like at least on the stuff you're working on in terms of like what we ship to customers.

Optimus production line.

So much like at least on the stuff you're working on in terms of like what we ship to customers.

Um,

Elon Musk: We've got the trees to prove it.

[Analyst] (Wolfe Research): Exactly. And then even on the optimistic side, right, as much as people think, "Okay, this is a new thing," I still remember, was it four plus years back, we were in a finance meeting with Elon, and Elon said, "Hey, our car is a robot on wheels." And that's where we started developing. In fact, most of the engineering team which is working on Optimus has come from the vehicle side. And that's why when we talk about manufacturing progress, we have the wherewithal because the same engineers who worked back in the day on drive units are working on actuators now. So that's where we can do if there is any company which can do it at scale, that is going to be us.

Vaibhav Taneja: Exactly. And then even on the optimistic side, right, as much as people think, "Okay, this is a new thing," I still remember, was it four plus years back, we were in a finance meeting with Elon, and Elon said, "Hey, our car is a robot on wheels." And that's where we started developing. In fact, most of the engineering team which is working on Optimus has come from the vehicle side. And that's why when we talk about manufacturing progress, we have the wherewithal because the same engineers who worked back in the day on drive units are working on actuators now. So that's where we can do if there is any company which can do it at scale, that is going to be us.

<unk>.

You know, hopefully with a production start towards the end of next year.

Mostly the same obviously customers, mostly just like them. They can choose the car wants to park in a spot or something that would just not super relevant for what actually but that's only a few minor things like those ones.

Mostly the same obviously customers like them. They can choose the car wants to park in a spot or something that would just not super relevant for what I see.

Um,

but that production ramp will take a while to get to an annualized rate of a million because

There's only a few minor things like those ones, but majority of the algorithms and architecture and everything is the same I think English about pumps.

It's going to move as fast as the slowest, dumbest, least lucky.

Majority of the algorithms and architecture and everything is the same I begin those pumps.

Thing out of 10,000 unique items.

Yeah, but I mean, as I mentioned earlier like we'll be adding reasoning to I don't know sure because that like reasoning and like $14. Three maybe 14 point for something like that.

Elon Musk: Ultimately, we will do Optimus 4. That will be 10 million units. Optimus 5, maybe 50 to 100 million units. I mean, it is really pretty nutty.

Yeah, but I mean, as I mentioned earlier like we'll be adding reasoning to I don't know sure because that like reasoning and like $14. Three maybe 14 point for something like that.

But it, but it will, it will get to a million units and then.

Ultimately, you know, we'll do optimist.

For that'll be, you know, 10 million units.

But I noticed you know for sure yeah, so with reasoning, it's literally going to think about which parking spot to pick.

But I know this unit for sure yeah, so with reasoning, it's literally going to think about which parking spot to pick at.

Optimus 5, Maybe.

50 to 100 million units. I mean, it's really pretty nutty.

Um,

At least I was going to say this is the entrance, but actually probably theres not pockets about right at the entrance if its a full.

Travis Axelrod: All righty. That is unfortunately all the time we have for Q&A today. Before we conclude, though, Vaibhav Taneja has some closing remarks.

yeah.

So it's gonna say this is the entrance, but actually probably theres not a parking spot right at the entrance if its a full you know if the if the parking lot is fairly full probably have open parking spot right at the entrance is very low.

Elon Musk: But we also have actually added a lot of new engineers as well to the team. So there's actually a lot of the credit for the Optimus engineering is actually also new engineers, many of them that are just out of college, actually. So the Optimus engineering team is a very talented engineering team. I'd say like, Wow, actually. And the Optimus reviews at this point are the engineering review, and then there's the manufacturing review being done simultaneously with an iterative loop between engineering design and manufacturing. Then we see we design something and we see like, Oh man, that's really difficult to make. We need to change that design to make it easier to manufacture. So we've made radical improvements to the design of Optimus while increasing the functionality but making it actually possible to manufacture. I'd say Optimus 2 is almost impossible to manufacture, frankly.

I'm talking about is fairly cool probably have open parking spot right at the entrance is very low but actually what else was simply due is dropping off at the entrance of the store and then go find a parking spot.

Vaibhav Taneja: Thanks, Travis. I want to take the time to talk about an extremely important vote, which is being held on November 6. The meeting will shape the future of Tesla, and we are asking you, as our shareholders, to support Elon's leadership through the two compensation proposals and the reelection of Ira, Kathleen, and Joe to the board. It is a team sport. Here at Tesla, the board is an integral part of the winning team. Shareholders are at the center of everything we do at Tesla. A special committee has laid out a compensation package. Like Elon said, we don't even want to call it a compensation package.

All righty. Um, that is unfortunately all the time we have for Q&A today. Uh, before we conclude, though, Bob has some closing remarks.

Thanks Davis.

But actually what else was simply due is dropping off at the entrance of the store and then go find a parking spot.

I want to take the time to talk about an extremely important vote, which is being held on November 6th.

But it's going to get very smart about figuring out of parking as well just go to spot figure out it's gonna spot empty spots much better than humans, because you got 360 degree vision.

But it's going to get very smart about figuring out of parking as well just go to spot.

It's gonna spot empty spots much better than humans, because you got 360 degree vision.

The meeting will shape the future of Tesla, and we are asking you, our shareholders, to support Elon's leadership.

Through the 2 compensation, proposals.

And it's gonna Yeah, Yeah, like I said just.

And it's gonna Yeah, yeah.

And the re-election of Ira Kathleen and your to the board.

Like I said just.

It's going to reuse the reasoning.

It's going to reuse the reasoning to solve things.

This whole things.

And putting that all inside your computer.

Note that it is a team sport, and here at Tesla, the board is an integral part of the winning team.

Okay.

Inside your computer.

Or is it extra shot and that's what the team is working on.

Or is it actually shot and that's what the team is working on because obviously you can do anything on a server that takes forever, but then car you need to make some decisions.

Anything you can do reasoning on the server that takes forever, but then car you need to make some decisions.

Shareholders are at the center of everything we do at Tesla, and a special committee has laid out.

Elon Musk: Yeah. The point is that I just, like, there needs to be enough voting control to give a strong influence, but not so much that I can't be fired if I go insane. You know, I think that sort of number is in the mid-20s, approximately. As a company that has already gone public, we've investigated every possible way to achieve increased voting control. Is there some way to have, like, a super voting stock? There really isn't. There is no way to have a super voting stock after you've gone public. For example, Google, Meta, many other companies have this, but they had it before they went public, and it sort of gets, I guess, grandfathered in. Tesla does not have that.

So putting all that into the computer that's in the car that's the challenge.

So picking on the computer and that's in the car that's the challenge.

Yeah, that's why I say like.

Pretty good understanding of like AI, that's sort of the giant model level with Croc and with Tesla and I'm confident in saying that Tesla has that that tells the AI has the highest intelligence density when you look at the intelligence per gigabyte.

Pretty good understanding of like AI, that's sort of the giant model level with Croc and with Tesla and like I'm confident in saying that Tesla has that tells the AI has the highest intelligence density when you look at the intelligence per gigabyte.

Compensation package. Like Elon said, we don't even want to call it the conversation tax. Yes, it's not the point. The idea is that there needs to be enough voting control to give a strong influence, but not so much that I can't be fired if I go insane. Um, but I...

Elon Musk: But my two bits point, we've gone from a person in a robot outfit to what people have seen with Optimus 2.5 where it's doing kung fu. It was like Optimus was at the Tron premiere doing kung fu just out in the open with Jared Leto. Nobody was controlling it. It was just doing kung fu with Jared Leto at the Tron premiere. You can see the videos online. And actually, the funny thing is a lot of people walked past it thinking it was just a person. Even though with Optimus 2.5, you can see that it has a waist that's three inches wide, which results in not a human. But the movements were so human-like that people didn't realize a lot of people didn't realize they were looking at a robot.

I think like Tesla is probably an order of magnitude better than anyone else.

I think like Tesla AI, it's probably an order of magnitude better than anyone else.

You know, and I I think that's sort of number is in the mid 20s approximately. Um, as a company that has already gone public, there's no that that we've investigated every possible way to. How do you achieve increased voting control without? Um,

You know.

And it isn't it doesn't have any choice because that has got to fit in the Apple computer.

um,

And it isn't it doesn't have any choice because that has got to fit in the Apple computer.

Hum.

The.

The discipline of having that level of AI.

The.

The discipline of having that level of AI.

Intelligence density.

if if there's some way to have like a super voting stock, but there there really isn't as there is no way to have a super voting stock after you've gone public. Um, but for example, uh, Google

Intelligence density.

We will pay great dividends. When you go to something that has an order of magnitude order of magnitude more capability like AI five.

Uh, meta. Um

You know, many other companies have this.

um,

Now you have that same intelligence density but.

Now you have that same intelligence density but no.

But you got 10 times more capability in the computer.

But you're about 10 times more capability in the computer.

Elon Musk: Like I said, I just don't feel comfortable building a robot army here and then being ousted because of some asinine recommendations from ISS and Glass Lewis, who have no frigging clue. I mean, those guys are corporate terrorists. Let me explain the core problem here. So many of the index funds, the passive funds, vote along the lines of whatever Glass Lewis and ISS recommend. They've made many terrible recommendations in the past that, if those recommendations had been followed, would have been extremely destructive to the future of the company. If you've got passive funds that essentially defer responsibility for the vote to Glass Lewis and ISS, then you can have extremely disastrous consequences for a publicly traded company if too much of the publicly traded company is controlled by index funds. It's de facto controlled by Glass Lewis and ISS.

Yeah.

Great and the next question will come from Collyn at Oppenheimer <unk> co.

But they had it before they went public, and so it sort of gets, I guess, grandfathered in. Um, Tesla does not have that. Um, so it's just.

Great.

Question will come from Collyn at Oppenheimer, Colin Please on mute yourself when you're ready.

Colin please on mute yourself when you're ready.

Like I said, I just don't feel comfortable building a robot army here and not, and then,

uh,

I'll go ahead, and let me yourself. Please thanks, so much guys.

I'll go ahead and limit yourself. Please thanks, so much guys.

You know, being ousted because of some S9 recommendations from ISS and Glass Lewis, who have no freaking clue.

I appreciate you, bringing up the challenges at hand, dexterity and humanoids, along with the complexity of the supply chain and the vertical integration you guys are pursuing yeah, I'm just trying to harmonize.

I mean, those guys are corporate terrorists.

And and the problem. Yeah. So let me like explain like the core problem here is that

Immunize the timeline for the start of production next year with the current state of the supply chain and what sounds like a fair amount of work remaining on the dexterity before you can really frees the hardware design and start to scale up production.

Elon Musk: What I'm saying is Optimus 3 will be a giant improvement on that and made at scale. Like I said, a very difficult thing. The Optimus sort of engineering and manufacturing reviews, and there's the Friday night meeting with Optimus, which sometimes goes till midnight. Then my Saturday meeting is the Saturday afternoon is with the AI5 chip design team. Those two things are crucial to the future of the company.

Uh, so many of the index funds, um, the passive funds.

Vote along the lines of whatever Glass Lewis and ISS.

It sounds like a fair amount of work remaining on the dexterity before you can really frees the hardware design and start to scale up production.

Recommend.

Hmm.

Now they have made many terrible recommendations in the past that, if those recommendations had been followed, would have been extremely destructive to the future of the company.

The hardware design will actually be frozen.

Well, we're not the hardware design will actually be frozen.

Even through startup production there'll be continued iteration.

Even through started production there'll be continued iteration.

Defer responsibility for the vote.

Because a bunch of the things that you discover are very difficult to make you only find out pretty late in the game. So we'll be doing rolling changes.

Uh, $2 billion, Glass Lewis and ISS.

Because a bunch of the things that you discover are very difficult to make you only find out pretty late in the game. So we'll be doing rolling changes.

For the Optimus design, even after sort of production but.

For the Optimus design, even after solder production.

But I do think that the.

Elon Musk: This is a fundamental problem for corporate governance because they're not voting along the lines that are actually good for shareholders. That's the big issue. That's what it comes down to: ISS, Glass Lewis, corporate terrorism.

You know the new hand is.

Um, then you can have extremely disastrous consequences for a publicly traded company. If too much of the publicly traded company is controlled by index funds, it's de facto controlled by Glass Lewis and ISS.

The new hand is.

And incredible piece of engineering and.

Hum.

This is a fundamental problem for corporate governance.

Travis Axelrod: Great. Dan, did you have a follow-up?

An incredible piece of engineering.

Daniel Roeska: Yeah. Just as a related, maybe you could just talk about to what extent are the AI efforts at Tesla and xAI complementary, or are they just different forms of AI? Maybe you can just help distinguish for the audience. Thank you.

And.

Yeah, that's yeah, well like I said, we will have a.

Yeah, that's yeah, well like I said, we will have.

Because they're not voting along the lines that are actually good for shareholders.

A production intent prototype.

To show off in Q1, and probably February or March.

That's the big issue. I mean,

Ready to ship.

So often.

Q1, probably February or March.

Vaibhav Taneja: Yeah. I would say, you know, the special committee did an amazing job in constructing this plan for the benefit of the shareholders. There's nothing which gets passed on till the time shareholders make substantial returns. That's why, you know, in the end, I would say, I would urge you to not only vote on the plan, but also vote on all the three directors because of their exceptional knowledge and experience. Literally, you know, we at Tesla work with these directors day in, day out. I mean, there's not even a single day that one of the directors I haven't spoken to or one of my colleagues hasn't spoken to. Even the directors out here are not just reading out of, you know, PowerPoint presentations. They're actually working with us day in, day out. I just urge you guys as shareholders to vote along the board's recommendation.

And then we're.

That's what it comes down to, uh, ISS, Glass Lewis, corporate terrorism.

And then we're.

Elon Musk: Yeah. They are different forms of AI. So the xAI, Grok is like a giant model that you could not possibly squeeze Grok onto a car, that's for sure. It is a giant beast of a model. With Grok, it's trying to solve for artificial general intelligence with a massive amount of AI training compute and inference compute. So for example, Grok 5 will actually only run effectively on a GB300. That's how much of a beast that Grok 5 is. So whereas Tesla's models are, I don't know, maybe about less than 10% the size, maybe closer to 5% the size of Grok. So yeah, they're really coming at the problem from very different angles. xAI and Grok, they're competing with Google Gemini, OpenAI ChatGPT, and that kind of thing. And some of it's complementary.

Yeah, we're gonna be building a million unit.

Optimists production line.

Yeah. I would say, you know, the special committee did an amazing job in constructing this plan for the benefit of the shareholders.

You know hopefully with the production start towards the end of next year.

There's nothing that gets passed down until the shareholders make substantial returns.

But that that production ramp will take awhile to get to an annualized rate of vermilion, because it's going to move as fast as the slowest dumbest least lucky.

But that that production ramp will take awhile to get to them.

Annualized rate of Vermilion, because it's going to move as fast as the slowest dumbest least lucky.

Thing out of 10000 unique items.

So that's why, you know, in the end, I would say I urge you to not only vote on the plan but also vote on all three directories because of the exceptional knowledge and experience. Literally, you know, we are Tesla and work with these directors.

But it will get to a million units.

But it will it will get to a million units.

And then I will.

Day in, day out. I mean, there's not even a single

And then I'll.

But we will do optimists.

Poor that'll be you know 10 million units.

Four that'll be you know 10 million units.

Optimists five maybe.

50 to 100 million units I mean, it's really pretty nutty.

Yeah.

Alrighty that is unfortunately, all the time, we have for Q&A today and before we conclude though that Bob has some closing remarks.

Vaibhav Taneja: Thank you, guys.

Travis Axelrod: Great. Thank you, Vaibhav. We appreciate everyone's questions today. We look forward to talking to you next quarter. Thank you very much and goodbye.

Day, that one of the directors, I haven't spoken to or one of my colleagues hasn't spoken to. And we're even the directors out here are not just reading out of, you know, PowerPoint presentations. They're actually working with us day in, day out. So again, I just urge you guys as shareholders to vote along with the board's recommendation. Thank you, guys.

Thanks Tavis.

Want to take the time to talk about an extremely important work, which is being held on November six.

I want to take the time to talk about an extremely important work, which is being held on November six.

Great. Thank you, Baba. We appreciate everyone's questions today. We look forward to talking to you next quarter. Thank you very much, and goodbye.

Meeting will shape, the future of Tesla and we are asking you as our shareholders to supporting lost leadership.

The meeting will shape, the future of Tesla and we are asking you.

As our shareholders to supporting lost leadership.

Two the two compensation proposals.

Two the two compensation proposals Andrew.

And the reelection of Idaho, Kathleen and Joel to the board.

And the reelection of Idaho, Kathleen and Joe to the board.

Elon Musk: For example, for Grok voice, being able to interact with Grok in the car is cool. For Optimus, voice recognition and voice generation is Grok. So that's helpful there. But they are coming at it from kind of opposite ends of the spectrum.

It is a team sport and head of Tesla. The board is an integral part of the winning team.

It is a team sport and hear that Tesla the board as an integral part of the winning team.

Shareholders.

Shareholders are.

At the center of everything we do at Tesla and then.

At the center of everything we do at Tesla.

The committee has laid out.

Compensation package like Elon said, no we don't even want to call. It a compensation thought yeah. Its just not at the point as I just like it needs to be enough voting control to give a strong influence, but not not so much that it can't be fired if I go insane.

Compensation package like Elon said, no we don't even want to call. It a compensation thought yeah. Its just not at the point is that just like there needs to be enough voting control to give a strong influence, but not not so much that it can't be fired if I go insane.

Travis Axelrod: All righty. Adam, let's give it another try. When you're ready, please unmute yourself for the next question. All righty. Unfortunately, still having audio issues. So we're going to move on to Walt from LightShed. Walt, please go ahead and unmute yourself.

But.

And I think that's sort of numbers in the mid Twenty's approximately.

As a company that has already gone public there's no.

We investigated every possible way to how do you achieve increased voting control without them.

Yeah.

Is there some way to have like a super voting stock, but there really isn't a there is no way to help us with voting stock.

Is there some way to have like a super voting stock, but there there really isn't a there is no way to help us with voting stock.

[Analyst] (LightShed): Can you hear me now?

Travis Axelrod: Yes.

After you have gone public but for example, Google matter.

[Analyst] (LightShed): Perfect. Thank you. Just getting back to Austin, if you can remove the safety driver at your end, is the limitation in the Bay Area just regulatory, or is it kind of the market-by-market learning process? And I guess similarly in the eight to 10 markets that you mentioned to get added, is the decision there to put a safety attendant in the passenger seat or the safety driver in? Is that your step-by-step process to opening up a market, or is it really just the regulation in the individual market?

After you have gone public but for example, Google matter.

There are many other companies have this.

But they had they had it before they went public and so it sort of gets I guess grandfather then.

Because it does not have that so.

So.

It's just.

Like I said I, just don't feel comfortable building robot army here and not and then.

You know being ousted because of some asinine recommendations from ISS and glass Lewis, who I have no freaking clue.

You know being ousted because of some asinine.

Recommendations from ISS and glass Lewis, who I have no freaking clue.

Elon Musk: Well, I think even if the regulators weren't making us do it, we'd still do that as the sort of right sort of cautious approach to a new market. So just to make sure that we're being paranoid about safety, I think it makes sense to have sort of either a safety driver or a safety occupant in the car when we first go to new markets just to confirm that there's not something we're missing. Because all it takes is like one in 10,000 trips to go wrong, and you've got an issue. So it's just to make sure, is there some peculiarity about a city, like a very difficult intersection or, I don't know, something that's an unexpected challenge in a city for that one in 10,000 situation.

I mean, those guys are corporate terrorists.

And the problem yet so let me explain like the core problem here is that.

And the problem. So let me explain like the core problem here is that.

So many of the index funds.

Passive funds.

Vote, along the lines of what whatever glass Lewis and ISS recommend.

Recommend.

Now they have made many terrible recommendations in the past that if those recommendations have been followed would've been extremely destructive to the future of the company.

But if you've got passive funds that essentially to defer.

But if you've got passive funds that essentially the deferral of responsibility for the vote.

Q3 2025 Tesla Inc Earnings Call

Request a DemoDemo

TSLA

Tesla

Earnings