Q2 2025 Taiwan Semiconductor Manufacturing Co Ltd Earnings Call
Taiwan Semiconductor Manufacturing Co Ltd
Go away. Told you.
Man.
The drill is over. ???
The drill is over.
Was.
I am Su Zhikai from the Corporate Relations Department of TSMC.
Welcome to TSMC's 2025 Q2 Investor Conference
Since this legal briefing will be simultaneously broadcasted to global investors, we will use English throughout. Thank you for your understanding.
Good afternoon everyone, and welcome to TSMC's second quarter, 2025 Ernie's Conference and Conference Call. This is Jeff Su, TSMC's Director of Investor Relations, and your host for today.
Speaker Change: Good afternoon, everyone and welcome to tsmc's. Second quarter 2025 earnings conference and conference call.
Today's event is being broadcast live through TSMC's website at www.tsmc.com where you can also download the earnings release materials if you are joining us through the conference call.
This is Jeff Sue, tsmp, director of investor relations, and your host for today.
today's event is being webcast live through tsmc's website at
Where you can also download the earnings release materials. If you are joining us through the conference call, you're dialing lines are in listen-only mode.
The format for today's event will be as follows. First, TSMC Senior Vice President and CFO , Mr. Wendell Huang, will summarize our operations in the second quarter 2025, followed by our guidance for the third quarter 2025.
Afterwards, Mr. Huang and TSMC's Chairman and CEO , Dr. C.C. Wei will jointly provide the
The format for today's event will be as follows first tsmc, senior vice president and CFO Mr. Wendell Huang will summarize our operations in the second quarter, 2025, followed by our guidance. For the third quarter, 2025
Then we will open both the floor and the line for the question and answer session.
Speaker Change: Provide the company's key messages.
Speaker Change: Then we will open both the floor and the line for the question and answer session.
As usual, I would like to remind everybody that today's discussions may contain forward looking statements that are subject to significant risk and uncertainties, which could cause actual results to different maternity from those contained in the forward looking statements.
Please refer to the Safe Harbor notice that appears on our press release.
Speaker Change: As usual, I would like to remind everybody that today's discussions may contain forward-looking statements that are subject to significant risks and uncertainties which are called which could cause actual results to differ materially from those contained. In the forward-looking statements,
please refer to the safe harbor. Notice that appears on our press release.
And now I would like to turn the microphone over to TSMC CFO , Mr. Wendell Huang for the summary of operations and the current quarter-guide.
Thank you, Jeff. Good afternoon, everyone. Thank you for joining us today. My presentation will start with financial highlights for the second quarter 2025.
Speaker Change: And now I would like to turn the microphone over to tsmc CFO Mr. Wendell hang for the summary of operations and the current quarter guidance.
After that, I will provide guidance for the third quarter of 2020.
Wendell Hang: Thank you, Jeff. Good afternoon everyone. Thank you for joining us today. My presentation will start with financial highlights for the second quarter 2025
The second quarter revenue increased 11.3% sequentially in NT. As our business was supported by strong demand for our industry leading 3 nanometer and 5 nanometer technologies, partially offset
after that, I will provide the guidance for the third quarter of 2025
In US dollar term, revenue increased 17.8% sequentially to 30.1 billion, and exceeded our second quarter
Second quarter of Revenue, increase 11.3% sequentially in NT as our business was supported by strong demand for our industry-leading 3, nanometer and 5. Nanometer Technologies. Partially offset by an unfavorable foreign exchange rate.
Gross margin decreased 0.2% percentage point sequentially to 58.6%
In US dollar term Revenue. Increase 17.8% sequentially to 30.1 billion and exceeded our second quarter got guidance.
Primarily due to an unfavorable foreign exchange rate and margin dilution from our overseas
Due to operating leverage, operating margin increased 1.1% to 49.6%
Gross margin decreased 2, percentage points sequentially to 58.6%, primarily due to an unfavorable foreign exchange rate and margin dilution from our overseas. Fabs partially balanced by higher capacity, utilization and cost Improvement efforts
Overall, our second quarter EPS was 15.36 NT, up 60.7% year over year, and ROE was 34.8%.
due to operating leverage operating margin increased 1.1, percentage points, sequentially to 49.6%,
Overall our second quarter EPS was 15.36 NT up 60.7% year-over-year and Roe was 34.8%.
Now let's move on to revenue by technology
Three nanometer process technology contributed 24% of wafer revenue in the second quarter.
Now, let's move on to revenue by technology.
While 5 nanometer and 7 nanometer accounted for 36% and 14% respect [inaudible]
Advanced Technologies defined as 7nm and below accounted for 74% of wavering.
3, nanometer process technology contributed to 24% of wafer Revenue in the second quarter. While 5, nanometer and 7, nanometer accounted for 36% and 14% respectively.
Advanced Technologies defined as 7 nanometer and Below accounted for 74% of wafer Revenue.
Moving on to Revenue Contribution by PLAT
HPC increased 14% quarter-over-quarter to account for 60% of our second quarter revenue.
Moving on to revenue contribution by platform.
Smartphone increased 7% to account for 27%
HPC increased 14% quarter over quarter to account for 60% of our second quarter Revenue.
IoT increased 14% to account for 5%
Automotive, Stay Flat, and Accountable Fiber
Smartphone increased 7% to account for 27% iot increased 14% to account for 5%.
and DCE increased 30% to account for one.
Automotive, stay flat and accounted for 5% and dce increased 30% to account for 1%.
Moving on to the balance
We ended the second quarter with cash and marketable securities of $2.6 trillion NT, or $90 billion US dollars.
Wendell Hang: Moving on, to the balance sheet.
On the liability side, current liabilities decreased by 22 billion NT, quarter over quarter, mainly due to the decrease of 38 billion in the crew, liabilities, and others.
We ended the second quarter with cash and marketable securities of 2.6 trillion NT or 90 billion US dollars.
The decrease in accrual liabilities and others was mainly due to the payment of income.
On the liability side, current liabilities decreased by 22 billion NT quarter over quarter mainly due to the decrease of 38 billion in the crew liabilities and others.
The decrease in ACR liabilities and others was mainly due to the payment of income tax.
On financial ratios, accounts receivable turnover dates decreased five days to 23 days.
The decrease in the country's receivable was mainly due to anti-dollar appreciation, as almost all of our country's receivables are in U.S. dollars.
Wendell Hang: Our financial ratios accounts receivable. Turnover days, decrease, 5 days to 23 days,
Days of inventory decreased 7 days to 76.
The decrease in accounts, receivable was mainly due to Auntie dollar appreciation, as almost all of our accounts receivables are in US dollars.
primarily due to higher N3 and N5 wafer ship.
Wendell Hang: Days of inventory. Decreased 7 days to 76 days.
Uh, primarily due to higher, M3 and M5 wafer shipments.
Regarding cash flow and KPX, during the second quarter, we generated about 497 billion NT in cash from operation.
Spent 297 billion in K-PACs and distributed 117 billion for third quarter 24 cash dividends.
Regarding cash flow and kex. During the second quarter, we generated about 497 billion NT in cash from operations.
Taking the unfavorable exchange rate into consideration.
Wendell Hang: Spend 297 billion in kex and distributed. 117 billion for third, quarter 2424 cash dividend.
Our cash balance decreased $30.3 billion to $2.36 trillion at the end of the quarter.
Wendell Hang: Taking the unfavorable exchange rate into consideration.
In US dollar terms, our second quarter capital expenditure is total 9.6 billion.
Our cash balance decreased 30.33 billion NT to 2.36 trillion at the end of the quarter.
Billion.
I've finished my financial summary. Now let's turn to our current court guidance.
Based on the current business outlook, we expect our third quarter revenue to be between $31.8 billion and $33 billion US dollars, which represents an 8% sequential increase or a 38% year-over-year increase at the mid-point.
I have finished my financial summary. Now, let's turn to our current quarter guidance based on the current business Outlook, we expect our third quarter Revenue to be between 31.8 billion and 333 billion US Dollars. Which represents an 8% sequential increase or a 38% year-over-year increase at the midpoint.
Based on the exchange rate assumption of 1 US dollar to 29 NT, Rose Margin is expected to be between 55.5% and 57.5%
Operating Margin Between 45.5% and 47.5%
Based on the exchange rate, Assumption of 1 US dollar to 29 NT. Rose margin is expected to to be between 55.5% and 57.5%.
In addition, we maintain our 2025 capital budget to be between $38 billion and $42 billion U.S.
Wendell Hang: Operating margin between 45.5% and 47.5%.
This concludes my financial presentation.
In addition, we maintain our 2025 capital budget to be between 38 billion and 42 billion US.
This concludes my financial presentation.
Now, let me turn to our keymess.
I will start by talking about our second quarter 25 and third quarter 25 profitability.
Now, let me turn to our key messages.
Compared to first quarter, our second quarter growth margin slightly decreased by 20 basis points, sequentially to 58.6%
I will start by talking about our second quarter, 25 and third quarter, 25 profit.
This was primarily due to an unfavorable foreign exchange rate and margin dilution from our overseas
Compared to first quarter, our second quarter gross margin slightly decreased to buy 20 basis points sequentially to 58.6%.
Partially offset by higher than expected overall capacity utilization and cost improvement that
This was primarily due to an unfavorable foreign exchange rate and margin dilution from our overseas Fabs.
Wendell Hang: Partially offset by higher than expected, overall, capacity, utilization and cost improvements.
Compared to the first quarter, for the exchange rate of $1 to $32.8ANT, the actual second quarter exchange rate was $1 to $31.05NT.
This created about 220 basis points, margin headwind, to our actual second quarter gross
Wendell Hang: Compared to the first quarter, foreign exchange rate of 1 dollar to 32.8 Aunt. The actual second quarter exchange rate was 1 dollar to 31.05 NT.
We also experience slightly more than 100 basis points impact from the ramp-up of our overseas
Wendell Hang: This created about 220 basis points. Margin headwind to our actual second quarter, gross margin.
We also experienced slightly more than 100 basis points. Impact from the ramp up of our overseas. Fabs mainly as the margin dilution from our Arizona Fab started to kick in,
We have just guided our third quarter-grows margin to decrease by 210 basis points to 56.5% at the midpoint.
Primarily due to the continual unfavorable foreign exchange rate and more pronounced dilution from overseas facts as we ramp up further in Kumamoto and Arizona.
Wendell Hang: we have just guided our third quarter of growth, margin to decrease by 210 basis points to 56.5% at the midpoint,
We continue to forecast the gross margin dilution from the ramp up of our overseas fabs in the next five years starting from 2025 to be between 2% to 3% every year in the early stages and widen to 3% to 4% in the latter stage.
Wendell Hang: Primarily due to the continual unfavorable foreign exchange rate and more pronounced dilution from overseas. Fabs as we ramped up further in Kumamoto and Arizona,
We continue to forecast, the gross margin dilution from the ramp up of our overseas Fabs in the next 5 years, starting from 2025.
Despite the higher cost of overseas faps, we will leverage our increasing size in Arizona and work on our operations to improve the cost.
Wendell Hang: To be between 2 to 3%. Every year in the early stages and widened to 3 to 4% in the latter stages,
We will also continue to work closely with our customers and suppliers to manage the imp-
Wendell Hang: Despite the higher cost of overseas, Fabs, we will leverage our increasing size in Arizona and work on our operations, to improve the cost structure.
Overall, with our fundamental competitive advantages of manufacturing technology leadership and large-scale production base, we expect TSMC to be the most efficient and cost-effective manufacturing in every region that we operate.
Wendell Hang: We will also continue to work closely with our customers and suppliers to manage the impact.
Overall without fundamental competitive advantages of manufacturing, technology leadership and large scale production base. We expect tsmc to be the most efficient and cost-effective manufacturer in every region that we operate.
Now, let me make some comments on the impact of foreign exchange trade on TSNC's revenue and profitability.
NT Dollar is the reporting currency of our financial state.
Now, let me make some comments on the impact of foreign exchange rate on tsmc's revenue and profitability.
Nearly all of our revenue is in US dollars, while about 75% of cost of goods sold is in NT.
Anti dollar is the reporting currency of our financial statements.
Therefore, fluctuations in the exchange rate between US dollar and NT will have a sizable impact to our reported revenue and gross profit mark.
Wendell Hang: Nearly all of our revenue is in US Dollars while about 75% of cost of goods. Sold is in NT.
The sensitivity of the revenue to dollar anti-exchange rate is nearly 100%. That is, every 1% appreciation of NT against US dollars will reduce our reported NT revenue by 1%.
Wendell Hang: Therefore fluctuations in the exchange rate between US dollar and NT, will have a sizable impact to our reported revenue and gross profit margin.
The sensitivity of our gross margin to the same 1% exchange rate change is about 40 basis.
Wendell Hang: The sensitivity of the revenue to Dollar NT. Exchange rate is nearly 100%. That is every 1% appreciation of NT against US Dollars will reduce our reported anti- Revenue by 1%
That is, eventy appreciate 1% against the dollar. Our growth margin will come down by about 40
Wendell Hang: The sensitivity of our gross margin to the same. 1% exchange rate. Change is about 40 basis points.
Wendell Hang: That is aventi appreciate 1% against the dollar. Our gross margin will come down by about 40 basis points.
Compared with our second quarter exchange rate guidance of $1 to $32.5 NT, provided on April 17th.
The NT dollar has appreciated by an average of about 4.4 percent sequentially, which negatively impacted our second quarter revenue by about 4.4 percent in NT and our growth margin by about 180 percent.
Compared with our second quarter exchange rate. Guidance of 1 dollar to 32.5 Mt provided on April the 17th
Wendell Hang: Appreciated by an average of about 4.4% sequentially.
For the third quarter of 25, based on the current exchange rate of $1 to $29 NT, the average NT dollar will appreciate by another 6.6%.
Which negatively impacted our second quarter Revenue by about 4.4% in NT and our gross margin by about 180 basis points.
Sequentially, which will negatively impact our third quarter revenue by 6.6 percent.
Wendell Hang: For third quarter of 25 based on the current exchange rate of 1 to 29 NT, the average NT dollar will appreciate by another 6.6%.
In NT, a reduced growth margin by about 260 bases.
Sequentially. Which would will negatively impact our third quarter Revenue by 6.6%.
As a reminder, six factors determined TSMC's profitably
Wendell Hang: In NT and reduce our growth margin by about 260 basis points.
Leadership technology developed in the ramp up, pricing, capacity utilization, cost reduction, technology mix, and foreign exchange rate which is not in our control.
Wendell Hang: As a reminder 6 factors determine tsmc's profitability.
Wendell Hang: Leadership technology development and ramp up pricing capacity utilization cost reduction technology mix and foreign exchange rate which is not in our control.
When the foreign exchange rate is unfavorable as it is currently, we will focus on the fundamentals of our business and lean on the other five factors to manage through it, and we have successfully done in the past.
Thus, even with the unfavorable foreign exchange rate, we believe a long-term growth margin of 53% and higher remains well achieved.
When the foreign exchange rate is unfavorable as it is currently, we will focus on the fundamentals of our business and lean on the other 5 factors to manage through it and we have successfully done in the past.
Now, let me turn the microphone over to C.
thus even with the unfavorable foreign exchange rate, we believe a long-term gross margin of 53% and higher remains well achievable
Thank you, window. Good afternoon, everyone. First.
Now, let me turn the microphone over to Cece.
CeCe: Thank you, window.
Good afternoon, everyone.
First.
Let me start with our near-term demand outlook. We concluded our second quarter with revenue
of our guidance in U.S. out of town, many due to continue to robust AI and HPC related
CeCe: Let me start with our near-term demand Outlook. We concluded our second quarter with revenue of 30.1 billion in US dollar.
Above our guidance in US dollar term.
Moving into the third quarter, 2025, we expect our businesses in the third quarter to be driven by strong demand for our leading-age process technology.
Mainly due to continued, robust Ai and HPC, related demand.
Looking into the second car of 2024
Moving into the third quarter 2025, we expect our business in the third quarter to be driven by strong demand for our Leading Edge process Technologies.
We have not seen any change in our customers' behavior so far.
CeCe: Looking into second car for of 2025.
However, we understand there are uncertainties and risks from the potential impact of terrorist policies, especially on consumer-related and the price sensitive in market segments.
we have not seen any change in our customers Behavior so far,
however, we understand there are uncertainties and risk.
While we observe rebate programs in China are stimulating some near-term demand upside, we believe this is a short term in nature and continue to expect a mild recovery in overall NGIA and market segment in 2025.
From the potential impact of tariff policies, especially on consumer related and the price sensitive in market segments.
CeCe: While we observe rebate program.
Having said that, we believe the demand for Semiconductor is very fundamental and will continue to be robust.
CeCe: In China are stimulating some near-term demand upside. We believe this is a short-term in nature and continue to expect a mild recovery in overall non Ai and market segments in 2025.
Recent developments are also positive to AI's long-term demand outro.
CeCe: Having said that we believe, the demand for Semiconductor is very fundamental and will continue to be robust.
The explosive growth in token volume demonstrates increasing AI model usage and adaption, which means more and more computation is needed leading to more leading and silicon demand. We also see AI demand continuing to be strong.
CeCe: Recent developments are also positive to Ai and long-term demand Outlook.
including the rising demand from sovereign AI.
CeCe: The explosive growth in token, volume demonstrate, increasing AI model, usage, and adoption which means more and more computation is needed. Leading to more leading and silicon demand. We also see AI demand continuing to be strong.
Therefore, we now expect our four-year 2025 revenue to increase by around 30%.
Including the rising demand from Sovereign AI.
in the U.S. data term, supported by Sean Demand for our inter-3D, 3 nanometer and 5 nanometer technologies, underpinned by the growth in our HPC platform.
Therefore, we now expect our 4 year 2025 Revenue to increase by around 30%.
and Mr. Uncertainties, we will remain mindful of the...
CeCe: In US dollar term supported by strong demand for our industry, leading, 3, nanometer and 5 nanometer Technologies. Underpinned by growing our HPC grade form.
Potential tariff related impairs.
CeCe: Mr. Uncertainties, we will remain mindful of the
and be prudent in our business planning going into second half, 2025 and 2020 years, who are continuing to embrace for the future makeup trend. We will also focus on the fundamentals of our business.
Potential Terror related impact.
Technology Leadership, Manufacturing Excellence, and Customer Trust to further strengthen our competitive
CeCe: And be prudent in our business planning, going into second half 2025 and 2026. While continuing to invest for the future. Make a trend, we are also focused on the fundamentals of our business.
CeCe: Technology leadership manufacturing excellence and customer trust to further. Strengthen our competitive position.
Next, let me talk about TSMC's global manufacturing footprint update.
All our overseas decisions are based on customer's needs, the value of some geographic flexibility and a necessary level of government support. This is also to maximize the value of our shareholders.
CeCe: Next, let me talk about the tsmc's, global manufacturing footprint update.
All our overseas decision are based on customers need
With a strong collaboration and support from our leading U.S. customers, and the U.S. Federal State and City Government, we announce our intention to invest a total of U.S. 165 billion in advanced semiconductor manufacturing in the United States.
Is also to maximize the value of our shareholders.
CeCe: With a strong collaboration.
CeCe: and support from our leading us, customers and the US
This expansion includes plans for six advance-a-way-for-manufacturing firms in Arizona, two advance-a-packaging firms and a major R&D center to support the strong multi-year demand from our customers.
CeCe: federal state and city government will announce Our intention to invest a total of us 165 billion in advanced semiconductor Manufacturing in the United States.
CeCe: This expansion includes plans for 6 advanced wafer manufacturing 5 in Arizona 2 Advanced packaging files.
CeCe: And a major arc Center.
Our first webbing, Arizona, has already successfully entered into high-volume production.
CeCe: To support the strong multi-year demand from our customers.
in 4Q-2024, utilizing N4 process technology with a yield comparable to our fat in Taiwan.
CeCe: Our first web in Arizona has already successfully entered into high volume production.
The construction of our second fire, which was utilized with nanometer process technology, is already complete.
CeCe: In 4 q, 2024 utilizing N4 process technology with a yield comparable to our Fab in Taiwan.
We are seeing strong interest from our leading U.S. cash
The construction of our second fire which were utilized 3. Nanometer process. Technology is already complete.
and are working on speeding up the volume production schedule by several quarters to support their needs.
Construction of our certifier, which were utilized two nanometer and the H-16 process technologies has already begun and we were looking to speeding up
We are seeing strong interest from our leading us, customers and are working on speeding up the volume production schedule by several quarters to support their needs.
CeCe: Construction of our certified, which were utilized.
Production schedule as well
2, nanometer and a 16 processes has already begun and we were looking to speeding up.
based on the strong air-related demand from our customers.
The production schedule as well.
Our Force Web, Wai Yudurai, Hiantu, and the A16 Kim Process Technology, and our fifth and sixth and fifth, Wai Yuzh, even more advanced technology.
CeCe: Based on the strong AI related demand from our customers.
CeCe: Our Force web were utilized.
The construction and ramp schedule for those who have been based on our customers' needs.
CeCe: Into and A6 process technology and our fifth and sixth Fair were use even more Advanced Technologies.
Our expansion plan was a neighbor TSMC to scale up to a geek out web.
CeCe: The construction and RAM schedule, for those files will be based on our customers need.
Craster in Arizona to support the needs of a leading-edge customer, a smartphone, AI and HPC application.
CeCe: Our expansion plan will enable tsmc to scale up to a giga web.
CeCe: Cluster in Arizona.
CeCe: To support the needs of our Leading Edge customers in smartphone.
We also plan to build two new advanced packaging facilities and establish an R&D center to complete the AI supply chain.
Ai and HPC applications.
After completion, around 30% of our two nanometer and more advances.
We also plan to build 2 new Advanced, packaging facilities. And establish an R&D Center to complete the AI supply chain.
CeCe: After completion.
Capacity will be located in Arizona, creating an independent, leading Semiconductor Manufacturing in the U.S.
CeCe: Around 30% of our 2 nanometer and more advances.
The TSMC work continued to play a critical and integral role in enabling our customers success.
Speaker Change: Capacity will be located in Arizona, creating an independent leading as semiconductor manufacturing cluster in you in the US.
will also maintain a key partner in the network of the U.S. semiconductor industry.
Speaker Change: This tsmc will continue to play a critical and integral role in enabling our customers success.
Next, in Japan, sent through the strong support from the Japan Central, Prefecture, and local government, our first specialty technology firm in Kumamoto has already started voting production in late 2024, with very good EU.
We are also maintain a key partner and enable of the US semiconductor industry.
Speaker Change: Next in Japan, thanks to the strong support from the Japan, Central prefecture and local government. Our for specialty, technology 5 in Kumamoto,
Speaker Change: Has already started voting production in late 2024 with very good yield.
The construction of our second spatial defecation
scheduled to start a later this year, subject to the readiness of the local infrastructure.
Speaker Change: The construction of our second, spatial defair.
The ramp schedule will be based on our customers' need and market conditions.
Speaker Change: Is scheduled to start a data this year subject to the Readiness of the local infrastructure.
In Europe , we have received strong commitment from the European Commission.
Speaker Change: The ram schedule will be based on our customers need and market conditions.
and the German Federal, State and City Government, and are pre-progressing smoothly with our plans to build a special technology firm in Tristan, Germany.
In Europe, we have received strong commitment from the European Commission.
The REM schedule was also based on our customers' need and market conditions.
and the German federal state and city governments and our progressing smoothly with our plans to build a specialty, technology file interested, Germany,
In Taiwan, we support from the Taiwan government, we plan to build an 11-way-for-manufacturing FAP, then four advanced packaging facilities over the next several years.
Speaker Change: The ram schedule will also based on our customers need and market conditions.
We are preparing for multiple phases of two nanometer waves in both Sinsu and Goushung South Park to support the strong structural demand from our customers.
Speaker Change: In Taiwan, we support from the Taiwan government, we plan to build 11 wafer manufacturing Fab and 4 Advanced packaging for City over the next several years.
Speaker Change: We are preparing for multiple phases of 2, nanometer Fab in both Shinju and gowon, size Park.
by expanding our global footprint, who are continuing to invest in Taiwan, TSMC can continue to be the trusted technology and capacity provider of the global IC industry for years to come, while delivering profitable goods for our shareholders.
To support the strong, structural demand for our customers.
Now let me talk about our N2 and A16 stages. Our N2 and A16 technologies did the industry in addressing the insatiable demand for energy efficient computing, and almost all the innovators are working with the agency.
Well, delivering profitable growth for our shareholders.
Speaker Change: Now, let me talk about our in to and a 16 status.
We expect a number of new tables for two nanometer technology in the first two years to be higher than both three nanometers, five nanometers in the first two years, fueled by both smartphone
Speaker Change: Our in2 and a 16 Technologies. Lead the industry in a tracing. The insensible demand for energy. Efficient Computing and almost all the innovators are working with tsmc.
Speaker Change: We expect the number of new tables for 2 nanometer technology in the first 2 years to be higher than both 3. Nanometer,
and to work deliver for known performance in power banning.
5 nanometer. India. First 2 years fewer by both smartphone and hpg applications.
with 10-15 speed improvement at the same power, or 22-30% power improvement at the same speed, a more than 15% chip density increase, as compared with the N3E.
Speaker Change: And to what deliver for node performance and power benefit.
N2 is aware on track for both production in the second half of 2025.
Speaker Change: We tend to 15 15 hour for 2230 power Improvement at the same speed, and more than 15% chip density, increase as compared with the n3e.
A scheduled with a ramp of a similar to N3.
Speaker Change: N2 is aware on track for boarding production in the second half of 2025.
With our strategy of continuous enhancement, we also introduce N2P
Speaker Change: A scheduled with a ram profile for a similar to N3.
As an extension of our N2 family, N2P features further performance and power benefits on top of N2, and volume production is scheduled for 2nd half of 2020.
We saw strategy of continuously enhancement. We also introduced n2p.
We also introduce a 16 feature in our best-in-class super power rail or SPR.
As an extension of our into family into P features of further performance and power benefits on top of N2 and volume production is scheduled for second half 2026.
Compared with the N2P, A16 provides a 48-10% speed improvement.
We also introduced a 16 feature in our basing class, super power rail, or SP,
at the same power, or 15-20% power improvement at the same speed, and additional 7-10% chip density.
Compared with a n2p. A 16 provides a, for the A2, 10% speed Improvement.
A16 is best suitable for specific HPC products.
at the same power or 15 to 20% power Improvement at the same speed and additional 7 to 10% chip density in
with compressed signal grout and dense power delivering land work. More information is on track for 2nd half, 20th year.
Speaker Change: A 16 is best suitable. Shooted for a specific. HPC products.
We believe N2, N2P, H16, and its derivatives will fuel our N2 family to be announced another large and long-lasting known for TSMC.
Speaker Change: With complex signal routes and dense power delivery in land work. Volume production is on track for second half 2026.
We believe in 2 N2 p a16 and is derivatives for a few hours into family to be announced another large and long lasting known for tsmc.
Finally, let me talk about our A14 stage.
featuring our second generation, Donald C. Transistor Structure, a 14-watt deliver another four-known stride from N2, with performance and power benefits to address the increasing structuring demand for high performance and energy efficient computing.
Finally, let me talk about our A14 status.
Speaker Change: Featuring our second generation. General Street transistor structure.
Compute with N2, A16 will provide 10-15 speed improvement at the same power, or 22-30% power improvement at the same speed.
Speaker Change: A 14 watt deliver. Another 4 known stride from N2 with performance and power benefits to address the increasing structural demand for high performance and energy efficient computing.
Compared with cn2.
Speaker Change: A 16 W provide 10 to 15 speed Improvement at the same power.
and about 20% cheap density gas.
All 20 to 30% power Improvement at the same speed.
A 14-technology development is on track and progressing well with device performance and your improvement on or ahead of schedule.
Speaker Change: And about 20% chip density gain.
Our a 14 technology development is on track and progressing. Well,
Boarding Production is scheduled for 2028. We will continue our strategy of continuous enhancement with A14, including a super power rail offering planned for 2029.
Speaker Change: With Device performance and ear Improvement on or ahead of schedule.
We believe A14 is derivative.
Boarding production is scheduled for 2028. We will continue our strategy of continuously and husband with A14 including a super power rail offering planned for 2029.
will further extend our technology leadership position and enable TSMC to capture the course
Speaker Change: We believe a 14 and its derivative.
Speaker Change: Were further. Extend our technology leadership position.
risk concluding our key message and thank you for your attention.
And enable tsmc to capture the growth opportunities were into the future.
Thank you, CC. Does conclude our prepared statements? So, before we begin the question and answer session, I would like to remind everybody to please limit your questions to it at time so that we can allow all the participants in the opportunity to ask their questions. Please press the star then one on your telephone keypad now. If at any time you'd like to remove yourself from the questioning queue, please press star then two. So now let's begin the question and answer session. I will take the first few questions from the floor then we'll flip and alternate to those on the line. Maybe we'll go left center and then write sort of a sequence. And we'll start here with Gokul Hariharan from JP Morgan.
This concluding, our key message and thank you for your attention.
Speaker Change: Thank you CC. This uh, does conclude our prepared statements. So before we begin the question and answer session, I would like to remind everybody to please let me your questions to 2 at a time so that we can allow, uh, all the participants and opportunity to ask their questions. Uh, questions will be taken both from the floor and from the call, should you wish to raise your question in Chinese? I will translate it to English before our management answers, your question. Uh, for those of you on the call, if you'd like to ask a question, please press the star, then 1 on your telephone keypad. Now, if at any time, you'd like to remove yourself from the questioning queue, please press star, then 2
Thanks, Jeff, and good afternoon, CC and Vendel. First question on demand. I think CC you mentioned data center AI demand definitely looks better than maybe three months back. Last quarter you also mentioned COVOS capacity will probably come into balance by 2026. Is that still our view or you think that the capacity now starts to look tighter?
Speaker Change: Left Center, and then, uh, right sort of a sequence. Uh, and we'll start, uh, here with a go call. Harry Harren from, uh, JP Morgan.
Second, I think you talked about on-device AI as a potential future driver. Are you seeing more development on the on-device AI part? Is it better compared to maybe three, six months back? And lastly, near term, your four Q looks like you're expecting revenue to decline. Is that based on what your customers are telling, especially on the consumer side? Or is it just the SMC being cautious and conservative in terms of the guidance?
Thanks Jeff. Um and good afternoon CC and Vel. Um first question on demand um I think uh CC you mentioned data center AI demand definitely looks better than maybe 3 months back. Um last quarter you also mentioned Co was capacity will probably come into balance by 2026. Is that still our view or you think that the capacity now starts to look tighter?
Speaker Change: Second, I think you talked about on device AI as a potential future driver. Are you seeing more development on the on device AI part? Uh, is it better compared to maybe 3 6 months back and lastly near-term? Your 4q looks like you're expecting Revenue to decline.
Speaker Change: is that based on what your customers are telling, especially on the consumer side, or is it just tsmc being
Okay, Gokul, thank you. Again, for the benefit of those, of course, here in person on the line, please allow me to summarize your questions. So maybe we'll take them one by one. His first question is on the demand, particularly data center in AI-related demand. We have CC said in his remarks, it is certainly still even stronger. So his question is about the advanced packaging and co-wash demand into 2026. How do we see the supply demand gap narrowing or becoming more balanced for co-wash the...
Speaker Change: Cautious and conservative, uh, in terms of the guidance.
Speaker Change: Okay. Uh, Google. Thank you again for the benefit of those. Uh, of course here in person. And on the line, please allow me to summarize your questions. So maybe we'll take them 1 by 1. Uh, his first, uh, question is on the demand particularly Data Center and AI related demand. Uh, as we as CC said, in his remarks, uh, it is certainly still, uh, even stronger. So his question is about, uh, the
Speaker Change: Advanced packaging and cold wash demand into 2026. Uh, how do we see the supply demand Gap narrowing, uh, or becoming more balanced for Co-Op specifically?
Gokul, the demand from the AI is getting stronger and stronger. If you pay attention to what the-
for treating companies, COC.
and so the makeup trend for the...
Speaker Change: Goku, the demand for the AI, getting stronger and stronger. If you pay attention to what, uh, uh, for treating companies, CEO said,
AI is continuing to be strong
and so the makeup trend for the, uh,
And so is the call.
So now we are again, we are in in in in in in
Speaker Change: AI is continue to be strong and so is the core worth.
In a mold, try to narrow the gap. I don't want to use the balance. The last time you guys are misunderstood, what I said is that it's a bad wording. So I want to say we try to narrow the gap.
and so now we are again we are in a in a
In a mode, try to narrow the Gap. I don't want to use the balance. The last time, you guys are misunderstood what I said, you thought is
Speaker Change: That wording. So I was say we try to narrow the gap.
So, momentum still there and bring here.
and then the second question or second part is on device or edge AI. Gokul wants to know how is the development of customers the work on on device AI compared to maybe three to six months ago? What is the interest or activity level and how do we see this?
Speaker Change: All right. So, momentum is still there and break healthy.
Okay. And then the second uh, question or second part is on on device or Edge AI. Google wants to know, uh, how is the development of customers the work on on device? AI compared to vary? Maybe 3 to 6 months ago? What is the interest or activity level? And how do we see this?
As the last time I said, I said that it takes one to two years for my customer to complete the jail.
New Design of the Product
The movement is still going, they are still continuing to...
Speaker Change: Uh, as the last time I see it, I say that it takes a 1 to 2 years for my customer to, uh, complete the um, um, new design of the product.
Speaker Change: The moment is still going.
Time goes by, I'll say, sir. [inaudible]
The number of units is actually mild, but in the size increase, we continue to see that, and the size increase by about 5 to 10% and that kind of trend continues.
Speaker Change: They are still continue to, uh, uh, as time goes by, as I said, um, the increase on the edge device, uh, the number of the units is actually mild but then the dice size increased. We continue to see that and the dice size increased by about 5 to 10% and that kind of trend continued.
So, you had to wait around-
Thank you very much for one year to see the whole.
Speaker Change: Okay, so uh you had to wait around.
Speaker Change: The probably 6 months or to 1 year to see a expiration.
Okay, and then go to the second question of the final part on the near term. I think Gokul, your question is with our third quarter guidance implies fourth quarter is there any particular reason or any comment that we want to make about the implied fourth quarter?
Speaker Change: Okay, and then uh, go to the second question of the final part on the near term. Uh, I think Google. Your question is with our third quarter guidance, uh, implies, uh, fourth quarter, is there any particular reason, uh, or any comment uh, that we want to make about the implied fourth quarter? Uh,
Speaker Change: Um, business momentum.
The above statements do not represent the position of this station.
or you are become conservative. [inaudible]
Speaker Change: I think you did not mention Goku's comments.
Speaker Change: Or you have become conservative.
That company is more real.
Speaker Change: That company is more.
Real.
Speaker Change: Um,
We are a company that achieves what we say and reaches high targets.
Speaker Change: We are.
So your calculation, I think Charlie also is nothing. So a lot of you is calculating our reported number so that you can easily see that our first quarter is decreased.
We are a company that what we say is what achieve and achieve the high Target.
Speaker Change: And so your calculation I think Charlie also is not in. So a lot of you is calculating, our reported numbers so that you can easily
Speaker Change: see that, our first quarter is,
We're taking to the consideration of the possible impact of Kerry.
Speaker Change: Decreasing.
and a lot of other answers.
Um, we take into the consideration of the possible impact of tariff.
So it become more conservative.
That's our current A.
Speaker Change: And a lot of other uncertainties, so we become more conservative.
But I guarantee you with our technology leadership
That's our current attitude.
and Excellent.
Guo Wengui, Member of Parliament
There are any opportunities, we work.
Leadership position and excellent uh Manufacturing.
If there are any opportunities.
Speaker Change: we will catch and
Also, please don't forget to share and give a like, and remember to subscribe to our channel to receive the latest updates.
Okay, thank you, CCC. Thank you, Gokul.
Speaker Change: Be expected that we are achieve our high-end Target. Okay.
Sorry, let's go one by one. Second question that may be Charlie Chan from Wilkinson
Okay, thank you CeCe, thank you Coco. Yeah. Uh sorry, let's go 1 by 1 uh, second question. That may be Charlie Chan from Morgan Stanley.
Thanks for taking my question, good afternoon, C.C. when done Jeff. So first of all, Congress are very strong results, especially on the gross margin side, a very good exclusion indeed. So my first question is really also on the gross margin, because the accumulated effect in place almost like 4.4% is the point. So it took big, took ignore. So when you essentially consider your 206 Welfare pricing, so called reflecting your value, would you consider this effect impact? And it's completely confident to keep your margin similar to DCS level. I feel like 53% is low bar, so just want to immediately high and see if that effect impact can be considered to reflect your value.
Okay, so Charlie's first question is on margin and I guess pricing, he notes obviously is one of those that the big move in the exchange rate and therefore a big impact to our profitability and gross margins. So his question is looking ahead to 2026, can we reflect or earn our value from the including the effects impact into the pricing and therefore what is our confidence level on the gross margin for next year? Can it keep a similar level as this?
Uh, thanks for taking my question, good afternoon, CC went on Jeff. So first of all Congress for really strong results and especially on the gross margin side, a very good execution indeed. Uh, so my first question is really also on the gross margin because the accumulated effect in pay is almost like 4.4% this points. I it's a 2, big to ignore. Uh, so when TNC consider your 2026, the word for pricing so called the reflecting your value. Would you consider this F FX impact and is coming confident to, uh, keep your margin similar to this year's label. I I, I feel like a 53% is a low power, uh, so just want to admit a little bit high and see if that the FX impact can be, uh, considered to, um, reflect to your value. Thank you. Okay, so Charlie's first question is on, uh, margin and I guess pricing, uh, he notes obviously is Windows 7.
uh,
Who let me assure you that, sir?
Speaker Change: big move in the exchange rate and therefore big impact to our profitability and gross margin. So his question is looking ahead to 2026. Uh can we reflect or earn our value uh from the including the FX impact into the pricing? And therefore, what is our confidence level on the gross margin for next year? Can it keep a similar level as this year?
Yes, the impaired form, the exchange rate is huge, but...
Speaker Change: well, let me assure you that, um,
Yes, the impact from the exchange rate is huge.
Speaker Change: Um,
You try to imply that whether we are…
Speaker Change: but,
Speaker Change: You try to imply that whether we are.
Speaker Change: um,
Until now
Speaker Change: still our value.
Let me answer that. We are working on this.
and we have confidence that the 53% growth margin in higher. I still want you guys.
Speaker Change: Let me answer that we are working on it.
Pay more attention to and the hype
Speaker Change: And we have confidence that the 533 gross margin and higher. I still want you guys to
Pay more attention to and higher.
Thank you.
Speaker Change: Okay.
Speaker Change: Thank you.
Hopefully, it will work out well. My second question is also very hot topic recently about the H20 chip shipping to China.
Speaker Change: Okay, thanks. Uh hopefully it will work out. Well uh my second question is also very hot topic recently about
I remember three months ago, there was a lot of questions on this matter, meaning that back then, I believe the chip was a suspend, but you're still very confident about your me 40% cager for Cloud Semiconductor in the coming five years.
The H20 chip, uh, shipping to China.
Speaker Change: Um, I remember 3 months ago, there was another question.
Speaker Change: On this matter, right? Meaning
Now, China becomes your address or market again. Do you think that the meat 40% keger target can be revised?
Speaker Change: Uh, that back then, I I I I believe the chip was suspended, but you still very confident about your M40 kegger for cloud semi growth in the coming 5 years.
Speaker Change: Right now, China becomes your address of Market. Again, do you think uh,
Thank you, Charlie. So, Charlie's second question is around the AI accelerator demand. He notes, of course, our customers' product, H20, recently, now seems to be able to ship to China, versus three months ago, it was not. So, his question really is, are long-term AI accelerator growth cager to grow close to mid-40s? Can it be higher? Do we think it will be higher? Is there upside?
That uh, uh me, 40% kegger Target can be revised off.
Speaker Change: Okay, thank you Charlie. So, Charlie's second question is, uh, around the AI accelerator demand, uh, he notes, of course, uh, our customers, uh, product age, 20 recently, uh, is now seems to be able to ship to China versus 3 months ago. It was not. Uh, so his question really is are long-term AI accelerator growth. Kegger to grow close to mid 40s. Uh, can it be higher? Do we think it will be higher? Is the upside to this?
Charlie, the age 20 now is again, according to the four trading companies CEO , we did not receive the signal yet, so it's too early to give you an estimate, but certainly it is a good news, right? I mean, that's a-
Speaker Change: Charlie. Uh,
Chen is a big market and my customer can...
Still continue to supply the chip to the big market, and it's a very positive
Speaker Change: The H20 now is uh according again, according to the World Trading companies, the CEO, you know, we did not receive the signal yet, so it's too early to give you an estimate, but certainly it is a good news, right? I mean that's uh, China is a big Market and my customer can uh,
and news for Dan.
Speaker Change: Still continue to supply, the chief to the big market, and it's a very positive.
and in return is a very positive news to DSM.
Speaker Change: News for them.
Whether we are ready to increase our forecast not yet, or down the quarter, probably will be more
Very helpful. Thanks. Thank you, Charlie. Okay, then we'll move on to the right side of the room for us. Bruce Liu from Goldman Sachs.
Thank you for thinking my question. I think Charlie already asked if there was a profit of your question already, so I just moved on to your end to read
Speaker Change: And in return is a very positive news to tsmc. Whether we are ready to, uh, increase our forecast, not yet. Uh, another quarter probably will be more appropriate to answer your question. Sure, answer your comment. It's a very helpful, thanks. Thank you Charlie. Uh okay. Then we'll move on uh to the right side of the room for us. Uh Bruce Liu from Goldman Sachs.
So, what's the revenue contribution? We can expect for the N2 ramp for Nick here. I'm a little surprised to hear that the N2 ramp is similar rate with N3, with N2, you do have both HPC and smartphone customer ranging out at the same stage, or in the first slide, year one or year two. Can we expect, like, you know, 15% revenue contribution coming from N2 for next year or a similar level compared to N2, which is around like 10, 11% for the year two.
Bruce Liu: Thank you for taking my question. I think Charlie already asked the profitability question already. So I just move on to your end to rent.
Okay, Bruce's first question is around the N2 ramp. His question is about the ramp and the ramp profile because we said the N2 ramp profile is similar to N3. So what does that mean? And then also, of course, what revenue contribution do we expect or can we share for N2 in 2026?
Bruce Liu: A little bit surprised that you hear that and 2 RAM is similar rate within 3 with until you do have like both HPC and and smartphone customer vending on the system stage uh or in the first like year 1 or year 2, right? Can we expect like, you know, 15% Revenue contribution coming from into for next year or or a similar level uh you know compared to to which is around like 10 11% for the year.
2. Okay, Bruce's first question is around the N2 ramp. Uh, he his question is about the ramp uh, and the ramp profile because we said the end to ramp profile is similar to N3. So what does that mean? And then also of course what uh Revenue contribution do we expect or can we share for into in 2026?
Bruce, I have a good argument.
Usually, we ramp up a new node using the smartphone. You knew that? Everybody knew it. Now, it's not only a smartphone but also a HPC product.
Bruce Liu: Bruce, you have a good uh, argument.
However, the ramping profile I just reported is similar to three nanometer. It's limited by our capability to build a new fire to ramp it up. And also a little bit is...
Speaker Change: Usually we ramp up a new node, using the smartphone you use that, everybody knows it. Now is not only a smartphone but also HPC product.
Speaker Change: Uh however the ramping profile. I just reported say similar to 3 nanometer. It's limited by our capability, to build a new Fab to ramp it up.
They also a little.
Street Forward is constrained by the campaign.
Speaker Change: Bit.
So, we say the ramp profile is similar to N3, but...
Speaker Change: Is straight forward is constrained by the capacity.
The revenue contribution certainly will be bigger because you don't expect our end-two is the same price as our end-three
so uh, we say the rain profile is similar to N3 but
Of course, good. Thank you. If that is the case, we should assume that in 27 and 2, Remba will be faster, right? Because you take 12 or 18 months for you to be a new fan, you should be able to achieve the even higher growth in until in 27.
Speaker Change: The revenue contribution certainly will be bigger because you don't expect our N2 is the same price as the N3, right?
So Bruce is asking if the revenue contribution is higher in 26, then should it be even greater in 27?
Of course, good, thank you. If if that is the case, you we we should assume like in 27. And 2 remember will be faster, right? Because you take 12, 18 months for you to put new fat, you should be able to, to achieve the even higher growth in until in 27 Wright.
We will answer the question in 26. Thank you.
So so Bruce is asking if the revenue contribution is much is higher in 26, then should it be even greater in 27?
We will answer that question in 26.
The next question is for N5 and N3. I want to understand the supply demand for N5 and N3 in the coming two years. Most of the AI will migrate to N3 next year, but it seems to me that the N5 conversion is mostly done in N5 and N3 conversion is mostly done. We don't really see green field capacity expansion from N3. It becomes super tightness for the coming years. So something that N5 will be lower utilized, or we try to build more N3 in the future, or what kind of, or we should see, we can sell out more value for N3 and N3.
Thank you.
Speaker Change: Okay. Uh, the next question is for M5 and 3, right? So uh I want to understand the supply demand for the M5 and 3 you know coming 2 years as also most of the AI with migrate to and 3 next year. But it seems to me that the the M5 conversion is mostly down. You know, uni on the conversion is mostly done and we don't really see like Green Field capacity expansion from N3. So it it becomes like super like, you know, tightness for the entity for the coming coming years. So does that mean that N5 will be, you know, low utilized or, you know, we try to
Okay, Bruce's second question is around N5 and N3. He wants to know what is the outlook the supply demand at these two advance notes becoming two years. His observation AI products will migrate to N3 and the N5N3 conversion is mostly done. So his question is will three nanometer supply be very tight the next few years and I think the last part therefore can we earn our value or price for that tightness. And then on the flip side, what about five nanometer will it become lower utilization?
Be on the more entry in the future, or what kind of or we should see like you know we can sell out more value for entry inside next year.
Bruce Liu: Okay, Bruce's second question is around our N5 and N3. He wants to know, what is the Outlook? The supply demand at these uh, 2 Advanced nodes, the coming 2 years. Uh his observation AI products will migrate to N3 and the N5 N3 conversion is mostly done. So his question is will 3 nanometer uh Supply be very tight the next few years? And I think the last part therefore uh can we um, earn our value or price for that tightness and then on the flip side, what about 5 nanometer will it become uh, lower utilization.
I'd like your comment on that we have to share our value because we're tight in N3 capacity. It will be continued for a couple of years, very tight.
In fact, N5 also very tight.
Bruce Liu: I'd like your comment on that. We have to share our value because of very tight. Uh in entry capacity, it will be continued for couple of years. Um, very tight
Bruce Liu: And in fact, uh, N5 also very tight.
The demand is high because of... [inaudible]
Bruce Liu: Uh,
Bruce Liu: the demand is high because of, uh,
A lot of AI products still in the four nanometer technology node, and they were transitioned to three nanometers probably in the next two years. So in meanwhile, N5 is still very tight in capacity, N3 even tighter. And so we are working hard. One of the TSMC's advantages, we have...
Bruce Liu: A lot of AI products still in the 4, nanometer technology node and there are transition to 3 nanometer probably in next 2 years. So in meanwhile N5 are still very tight in capacity then 3, even Tighter and so we are working hard uh 1 of the tsmc's, uh,
How many countries Fox
and so we have between N7 and 5 and 3, even the future N2, we have a
Almost for each note, we have about 85-90% coming to.
Bruce Liu: Advantage is that we have uh, kick off a cluster. And so we have uh between N7 and 5 N3, even the future N2 we have
So, it's not free, but it's much easier for THMC to adjust or convert the capacity between those known and today let me share with you we are using the N7 capacity to support N5 because the N5 is too tight and then we are converting N5 to N3 as you just pointed out we are continuing to do that and so today we are out leading as technology is a capacity we define N7 and pillow are all very tight
Bruce Liu: A almost for each node. We have about uh 85 to 90% company tools.
Bruce Liu: And today let me share with you. We are using the N7 capacitive support in 5 because the N5 is too tight and then we are converting N5 to N3 as you just pointed out. We will continue to do that. And so uh, today we are our Leading Edge. Technology is a capacity.
Bruce Liu: Uh, we Define N7 and below.
Bruce Liu: Are all very tight.
Seeing that, we are working very hard to, again, using my, my sentence, narrow the gap between the demand and the gap.
Bruce Liu: and uh,
Bruce Liu: Since that, we are working very hard to, uh, again, using my, my sentence narrow, the gap between the demand and the capacity.
Okay, thank you, Bruce. Let's go to the participants online. We'll maybe we'll take two questions from the online and then we'll come back to the floor. Thank you.
Bruce Liu: Okay, thank you Bruce. Uh, let's go, uh, to the participants online. Uh, we'll maybe we'll take 2 questions from the online and then we'll come back to the floor. Thank you.
Bruce Liu: Operator.
He has now asked in question, Brett Simpson, irritate [inaudible]
Speaker Change: Yes. Now asking question, Brett Simpson irate
Thank you very much, I have a question about Wendell's currency reduction. You have proposed some suggestions and measures for currency reduction, but my question is, some top projects like FX and overseas FAB circulation are more structural currency reductions, trying to trigger more FAB efficiency. Can you give us some examples of real growth seen in your business structure? Have we already started to see digital currency reductions benefiting from AI efficiency? Thank you. Okay, Brett. Brett's first question is a bit involved. We are looking at the advantages of currency reduction. He noted unfavorable exchange rates and overseas vulnerabilities due to high prices. These are structural leads. So his question is, how can we capture our value or adjust our currency reduction prices to help reduce these? Also, we talked before about using AI in our operations. How much advantage do we get from currency reduction, for example, with our customers and other examples of using AI to help our pricing structure, and can we quantify it?
Yeah, thanks very much. Um, I had a question for Wendell on Gross margins and, um, it's always helpful. You've, you've laid a framework for some of the puts and takes to tsmc's. Gross margins. But my question is really um some of these headwinds like FX and the dilution from overseas, Fabs are more structural uh cost increases and to what extent can tsmc uh adjust way for
For pricing higher to neutralize these cost increases in your business and I guess, secondly, on this point, how much economic benefits are you, um, seeing, uh, from applying AI across the Fabs? Um, I mean, I think Nvidia has mentioned that they're working with tsmc closely strategically in areas like computational lithography um to try to drive further Fab efficiencies. So can you maybe just give us some examples where you're seeing real gains in your cost structure and um,
Speaker Change: Are we at a point where you're starting to see several points of growth margin benefit from AI efficiencies? Thanks. All right. Brett. So, uh, Brett's first question is, uh, a little bit, uh, involved, but looking at our gross margin and profitability, uh, he notes that the unfavorable exchange rate in the dilution from the overseas due to the higher costs, these are structural headwinds. So his question is, can we? Uh, how can we, or can we, uh, earn our value or adjust, our wafer price to help uh, offset some of these? And also, how much, uh, we've talked about before about using AI ourselves in our operations. How much economic benefit are we deriving from things such as CEO, lethal with our customers such as uh, and are there other examples of using AI where it's helping our cost structure and can we quantify
Quantified the Benefit, sorry. That's your question, right? Right? [inaudible]
That's right, thanks Jeff.
Speaker Change: That the quantify the benefits. Sorry, that's your question, right? Right.
Okay, Brett, the first question, Gross Margin.
Speaker Change: That's right. Thanks. Thanks Jeff.
That's the reason why we've been talking about these six factors affecting our profitability.
Okay. Brad uh the first question, uh, gross margin.
I don't think I need to repeat those six factors, but whenever, for example, using foreign exchange rate, for example, a few years ago, there were also periods of time that foreign
Speaker Change: Uh, that's the reason why we've been talking about the 6 factors affecting our profitability.
So we are able to lean on the other facts.
to help us mitigate the negative impact from certain factors and therefore still achieve our growth margin targets. And you specifically asked about ASP, raising the price, but the price is just one of the
and I believe CG just elaborates a lot on earning our value, and at the same time there are other factors that we can leverage on. So all you know, that's why we're saying 53% and higher and higher gross margin is still achievable.
Your second question, though, AI benefits.
Speaker Change: Uh, the I I don't think I need to, uh, repeat those 6 factors. Uh, but whenever, for example, uh, using foreign exchange rate, uh, example, a few years ago, there were also periods of time, the foreign exchanges were against us. Uh, so we are able to lean on the other factors, uh, to help us, uh, mitigate the negative impact from certain factors and therefore still achieve. Our gross margin targets and you specifically ask about, uh, ASP raising the price, but the price is just 1 of the factors. Uh, and I believe, uh, CC just, uh, elaborate a lot, uh, on earning our value. And at the same time, there are other factors that we can leverage on. So all, you know, uh, that's why we're saying in 53% and higher and higher, uh, gross margin is still achievable.
I think we also talked about that before. We used that in operation, in manufacturing. We also used that in R&Ds. And just think about if we are able to produce 1% of productivity gains in the company of our size, that equals to $1 billion US dollars.
Speaker Change: Uh, your second question on AI benefits.
So that's the number we can share with you without going into too much others.
Uh, I think we also uh, talked about that before we use that in operation uh, in manufacturing. We also use that in r&d's. Uh and just think about, if we are able to produce 1% of productivity gains in the company of our size that equals to uh, 1 billion US dollars.
Speaker Change: Uh so that's the number we can share with you uh without going into too much other details.
Does that answer your question, Brett? Okay, yeah.
This is great, thank you very much, Wendell. I guess my key question is, I guess in your prepared remarks you mentioned that Taiwan has 11 fabs. I guess you are still planning 8 fabs abroad that have not yet been launched in the market. Can you talk about it? I have never seen a roadmap like TSMC's, it's huge. Can you share with us if you plan to have a bigger new energy expansion next year? I say this because in the past few months, we have seen a lot of GW data center announcements. I guess this week we have an important one from META. So the demand seems very strong. I'm just wondering if you have enough capacity to meet next year's demand. You plan between 5nm to 3nm, what are your thoughts on the N2 energy plan for 2026? Thank you. Brett's second question, he mentioned that we are in Taiwan and also abroad. He has never seen TSMC's energy expansion. He also noted that the demand for data centers remains very strong. His question is simple: do we have enough capacity to support next year, especially 2nm? Will we continue to transition from 5nm to 3nm? That's all his questions.
Does that answer your question? Brett? Okay. Yeah.
Speaker Change: Yeah, that's uh, that that's great. Thanks. Uh, thanks very much Wendall. Um, I guess my my follow-up question. Um, I guess just uh, digesting your prepared remarks CC. You mentioned 11 fibs in Taiwan, I think I count 8 fabulous that you're planning. Uh, that aren't commercially online yet. Um, can you maybe just talk a bit about? Uh, I mean, I've never seen that type of uh, construct. You know, like that that type of road map before uh, from tsmc it's it's it's quite big. Um, can you maybe share with us if you're planning a bigger expansion of new capacity next year? And I say this because, you know, in the last few months, we've seen so many gigawatt, uh, data center announcements, um, I think this week we had 1 from Neta that was significant. Um, so I I are you, uh, the demand looks very strong and I'm just wondering whether you have enough capacity to satisfy demand, uh, next year. Um,
Speaker Change: Whether you plan to convert further 5, nanometer to 3 nanometer. Um, and uh and how how you how you see the the uh N2 capacity um, plans for 2026. Thank you.
Brett? Thank you. Your observation is right.
Speaker Change: Okay. Uh, Brett second question. Uh, he notes that we are building uh uh many Fabs, both in Taiwan and also overseas. Uh, he's never seen this uh, size or scale of uh, capacity extension from tsmc before. So it's very and he also notes that the demand, you know, from data centers continues to be very strong. So, his question is basically very, simply do we have enough capacity, uh, to support, uh, the strong demand, uh, specific to next year, and also very specifically, to 2, nanometer and will we further? Also convert more 5, nanometers to 3? That's, I think, uh, all of his question.
Recently, we saw a lot of announcement of the AI data center of the world.
Speaker Change: Correct, thank you. You are observation is right? Um
and the demand on 3 nanometer, actually on 5 nanometers, 3 nanometers, and the future of 2 nanometers are very high.
Speaker Change: Recently, we saw a lot of announcement of the AI data center all over the world.
Speaker Change: And the demand on Sweden, nanometer on actually, on 5 nanometers 3 nanometers and the future of 2 nanometer.
and we did not see this kind of a strong demand for a long time, but...
Are very high.
Speaker Change: And uh, we did not see this kind of a strong demand.
Speaker Change: For a long time. But uh, uh,
What be enough to support them? I still want to use my word, say that we try very hard to never get between the demand and the support.
Speaker Change: What be enough to support them? I still want to use my word say that we try very hard to narrow the gap.
We're working very hard.
Between the demand. And the supply, we're working very hard.
Okay, thank you, Brett. Let's go to the next participant on the call.
Okay, thank you Brett. Uh, let's go to the next participant on the call.
with Arthur Lai from Korea Light is Open Now.
Speaker Change: Now, it's out the line from a query line is open now.
Hi, thank you, CCE Wendell and Jeff, Arsalan from Macquarie. Again, congrats on a strong result. I would like to follow up on the N2. I think, as CCE highlighted, this is a very exciting note. We are all heard. And then I want to follow up on the return on investment. Can you compare to the N2 and N3, the return on investment and then give us more color? Second one is, the reason we ask this question is because the K-test per area actually N2 is higher, right? And then we also heard from industry that the company's yield on the N2 is also pretty good. So can you give us some put and take how we think of the N2's future development?
Hi. Uh, thank you, uh, CC window and just, uh, on the query again. Uh, congratulations. A strong result. Uh, I would like, to follow up on the, on 2, I think. Uh, uh, as a cc highlight, uh, this is a very, uh, you know, exciting. Uh, note, we are all hurt. Um, and then I want to follow up on the return on investment. Uh, can you compare to the N2 and N3, uh, the return on investment? Uh, and then give us more color. Uh, second 1 is, uh, the reason we asked, uh, this, uh, question is because, uh, uh, the c c test, uh, the area actually, unto is higher, right? And then the, we also, uh, heard from the industry, like, the, the, um, companies that yield on the N2 is also, uh, pretty good. So, um, can you give us some potent? Take, uh, on how we think of the, uh, un tools?
Okay, so are there questions? Both questions I think are around N2. N2, as he said, is a very exciting note. He wonder who would like to know, understand what is the return or the return on investment that we see from N2 compared to N3. And also, can we talk a little bit into the capex is higher, but the yield is still very good. What is the developments that we're seeing for 2 nanometer? I think that is that your question, Arthur?
Speaker Change: The future development. Thank you.
Speaker Change: Uh, okay, so, uh, Arthur's question, uh, both questions. I think are around N2, uh, N2, as he said, is a very exciting note. Uh, he wonder who would like to know, understand, uh, what is the return or the return on investment, uh, that we see from into, uh, compared to N3 and also, um, can we talk a little bit into the capex, uh, is higher but the yield is still very good. Uh, What is the developments that we're seeing for 2, nanometer? I think that's. Is that your question are Arthur?
Yes, yes, yeah, etc.
Yes, yes, yes. Yeah. Exactly.
Okay, after N2 return, as we said before, [inaudible]
okay Arthur, uh, N2 return as we said before, uh,
N2, the profitability is better than N3.
Now there were questions asking how many quarters to catch up with the corporate before and N3 was it took N3 longer.
Speaker Change: And to, uh, the profitability is better than N3.
But for N2, we think it will be back to the old
Speaker Change: is asking how many quarters, uh, to catch up with the the corporate before and N3 was uh, it took N3 longer,
Having said that, I need to remind everyone that in the old days, we are talking about corporate average of, say, 50% gross margin. Nowadays, we're talking about 53% gross.
uh, but for N2, we think it will be back to the old days.
So it becomes less meaningful to talk about how much time it takes to catch up with corporate nowadays. But having said that structurally, N2 does have a better profitability day.
Uh, having said that I I need to remind everyone. That in the old days we were talking about corporate average of say, 50% gross margin nowadays, we're talking about 53% growth,
Speaker Change: Margin. So it becomes, uh, less meaningful to talk about how much time it takes to catch up with corporate.
An end-to-development is right on track. We're ramping it. In the second half of this year, we expect the revenue to come up in the first half of next year.
Speaker Change: Nowadays. But having said that structurally and to does have a better profitability than 3
Speaker Change: Okay. And then to development is right on track, we're ramping it uh in the second half of this year. We expect the revenue to come up in the first half of next year.
Okay, thank you, Wendell, thank you, Arthur, let's come back to the floor, we'll go left middle.
Right, so maybe Sunny Lin from UBIRD
Okay, thank you Wendell. Thank you. Uh Arthur. Let's come back to the floor, we'll go left middle. Uh
Speaker Change: Right. Uh, so maybe Sunny uh, Lynn from Ubi or
Good afternoon. Thank you for taking my question. Very good results and congrats. So my first question is followed on K-Paks. So obviously, four years out, guidance is stronger. You are turning more constructive on high-performance compute in AI. Yeah, you are keeping your K-Paks guidance. So it's fair to assume that you are considering some conservatism for K-Paks. For this year, give us an ongoing macro uncertainty, or is it because in the short-term your capacity expansion is somewhat constrained by the ability that you can wrap up capacity. And therefore, maybe in 2026 and 2027, I wish you expect some accelerations of your K-Paks guidance.
Lynn: Good afternoon. Thank you for taking my question. Uh, very good results and, uh, congrats. Uh, what's on my first question is to follow up on, uh, kex. So obviously, uh, 4 years, South guidance is stronger. Uh, you are turning more constructive on high performance computer and AI.
Okay, so Sunny's first question is around KPEC. She notes, of course, that we raised our 2025 revenue guidance this year, and we certainly still see a very robust demand from AI, yet we kept our KPEC's guidance in the same range of 38 to 42. So she wants to understand why is it because of macro uncertainty? Is it because of constraints in the construction and her other part of this question is what does the KPEC's outlook for? 26 and 27, I guess?
Lynn: I, uh, yeah, you are keeping your keypad guidance. Uh, so, uh, is it fair to assume that you are considering some conservatism, uh, for keypads for this year? Given the ongoing macro uncertainty, or is it because in the short term, uh, Your Capacity expansion is somewhat constrained, uh, by the ability that you can wrap up more capacity and therefore maybe in 2026 and 2027. I wish you expect some accelerations of your keypad spending
Lynn: Okay, so Sun's first question is around kex she notes? The course that, uh, we uh, raised our 2025, uh, Revenue guidance this year and we certainly still see a very robust demand from AI yet. We keep K, kept our capex guidance in the same range of 38 to 42. So she wants to understand. Why is it because of macro uncertainty, is it because of constraints in the construction and her other part of this question is, is what is the capex outlook for 26 and 27? I guess.
Sunny, the K-Packs, as we said before, the K-Packs invested in the given year is for the business opportunities in the following.
And as long as there are business opportunities, we will not hesitate.
to invent
Having said that nowadays, as CC also said, with all these macro uncertainties, we are mindful of these uncertainties, so we also take that into consideration in our capacity and capability.
Lynn: Okay. Uh Sunny uh, the kex as as we said, before the kex invested in a given year is for the business opportunities in the following years and as long as there are business opportunities, we will not hesitate uh, to invest.
Going forward, [inaudible]
Uh having said that nowadays uh at CC also said, with all these macro uncertainties, we are mindful of these uncertainties. So we also take that into consideration in our capacity and capex plan.
It's too early to talk about future years, KPACs, but I can share with you a company of our size.
Lynn: Uh, going forward.
Unliked that you see K-Pack's dollar amount suddenly drop a lot in any given years. That's all I-
Lynn: It's too early to talk about future years capex but I I can share with you a company of our size. It's
Lynn: unlikely that you see capex dollar amount suddenly drop a lot in any given years.
got it. Sounds like K-PAC could be going higher in the coming years. My second question is on Call AI, and so you think like earlier attribute most of the sales apply for 2025 to call AI. And therefore I wonder if you have an update on the Call AI growth in 2025, which you've got it before to be about 100% for 2025. And the implication to your co-risk capacity expansion, would you be, are you able to maybe expand a bit more co-risk capacity for this year to support a stronger call AI growth of this year? And any early insights that you could share with us for your co-risk capacity expansion for 2026. Thank you. Okay, so Sunny's second question is asking about our AI, well she says Cloud AI, basically our AI accelerator growth in 2025 and related the co-risk capacity. So her question is what is the AI accelerator revenue growth? We expect in 2025 and then what is our co-risk capacity expansion plan for 2025? And she asked a similar question to Charlie, or someone earlier, what is the plan for co-risk capacity in 2026?
Lynn: That's all I can share with you.
Speaker Change: Got it. Uh, sounds like, uh, kex could be going higher, uh, in the coming years. Uh, my second question is on cloud AI. Uh, and so you, uh, things like earlier achieve you, most of the sales apply for 2025 to Cloud Ai. And therefore, I wonder, uh, if you have an update on the call AI growth, uh, in 2025, would you guide it before, uh, to be about 100%, uh, for 2025, uh, and the implication, uh, to your co-worker price expansion? Would you be, uh, are you able to maybe expand a bit more correct capacity for this year, to Stronger, to support the stronger Cloud, AI growth, uh, for this year and any, uh, other early insights that you could share with us, uh, for your Co-op capacity expansion for 2026? Thank you, okay? So Sunny second question is asking about, uh, our AI. Well, she says Cloud AI basically our AI accelerator growth. Uh, in
Speaker Change: 2025 and related the kowas capacity. So her question is, uh, what is the AI accelerator Revenue growth? We expect in 2025 and then what is our Co-op capacity expansion plan for 2025? And she asked a similar question to Charlie or someone earlier with this plan for kowas capacity in 2026.
Well, my answer stayed the same. We are trying very hard to narrow the gap.
Speaker Change: Well.
So stay the same. We are trying very hard to narrow the gap.
for now and for 2026
Uh,
Speaker Change: for now, you know, for 2026.
The demand, the momentum, are very hearsay and very strong. And so we are building many new facilities in the back end to increase the AI, increase the cold war schedule.
Speaker Change: The demand the momentum are very healthy.
Speaker Change: And very strong.
to support our customers.
Speaker Change: And so, uh, we we are building many new, uh, facilities in the back end to increase the AI capacity of increase, the core work capacity to support our customer.
AI DEMAND RATIO
and saw the co-host's capacity, the demand of Rick.
uh, AI demanded restaurant and so
Speaker Change: the core capacity, the demand is very strong.
Okay, thank you, Sunny. Then we'll move on to Lower Chen from City.
Okay, thank you. Uh Sunny then we'll move on to our lower Chen from City.
Thank you for taking my question and congrats for the good results and I'll look. My question is also about the AI chip. You mentioned that AI chip is getting bigger and bigger and also the power consumption is getting much more. So I'm just wondering that amount like your advanced technology, including like the advanced node, we also noted that during the symposium, TSMC also announced some of the new technology in advanced packaging as well. So I'm just wondering how do you kind of prioritize your leading age advanced packaging. During the symposium, we see that the system now wafer that kind of a new design. Do you have any like a plan or like timeline for the new technology? And do we think about that that should be kind of aligned with our most advanced node process like an N2 or an X16 going forward?
Lower Chen: Okay. Uh, thank you for taking my question and congrats for the good result and Outlook. Uh, my question is also about the AI chip CC. You mentioned that AI chip is getting bigger and bigger. And also, uh, the power consumption is, uh, getting much more. So, I'm just wondering that amount like your advanced technology including like the advanced node. Uh, we are also noted that during the Symposium, uh, tsmc also announced some of the new technology in advanced packaging as well. So, I'm just wondering, uh, how do you kind of prioritize your, uh, Leading Age, Advanced packaging? Um, during the Symposium, we see that the system now wafer, that kind of a new design. Uh, do you have any like, uh, uh, plan or like timeline?
So, Laura's first question is on events packaging, notes AID, die sizes are increasing, the need for power consumption or energy efficiency is rising. So, she wants to understand how our strategy for events packaging along with the advanced node development. Are there any specific packaging solutions that we're prioritizing? What about the timeline and roadmap? How does that match up with our advanced node road?
For the new technology and should we think about that? Uh that should be kind of aligned with our most advanced know uh process like a N2 or uh x16 going forward.
Nora, I think...
Speaker Change: Okay, so Laura's first. Question is, uh, on Advanced packaging notes. AI the die sizes are increasing, the need for power consumption, or Energy Efficiency is rising. So, she wants to understand how our strategy for advanced packaging, uh, along with the advanced node development. Are there any specific packaging solutions that were prioritizing? Uh, what about the timeline and road map? How does that match up with our, um, Advanced node, uh, road map?
TSMC's philosophy to develop technology is working with customer. The customer has such knowledge, but we develop the technology we increase the capacity for that.
Norah, uh, I think, um,
Tsmc is a philosophy to develop a technology is working with customer. The customer has such demand, we developed the technology, we increase the capacity for them.
Speaker Change: so, um,
Every customer is important to TSM
Speaker Change: priorities.
So, in the advanced packaging side,
A lot of customers using the different approach.
Every customer is important to tsmc. So uh, and in the advanced packaging side
So we are developing a variety of different packaging and advanced technology for all the customers.
Speaker Change: A lot of customer using the different approaches.
Speaker Change: So uh, we are developing a variety of uh different uh, backend packaging, advanced technology for all the
is related to the advanced leading-edge technology. The answer is yes. So we have a system integration, we have a cold-war era. That's a terminology. We have a lot of different names that I cannot
Speaker Change: uh,
whether is a related to the advanced Leading Edge technology. The answer is yes.
Speaker Change: Okay, so we have a system integration. We have a cold Wars error. That's a terminology. We have a lot of different name that I cannot even remember, but, uh, it's a lot of, uh, uh, varieties and we work with our customer to meet their demand that I can answer you.
It's easy to kind of leverage your transfer different kind of technology from Europe .
Speaker Change: Is that easy to kind of Leverage your transfer different kind of Technology from your perspective.
So, Laura is asking how fungible are these different packaging technologies? How interchangeable or easy to transfer the technology between different packaging solutions? Of course, there are some similarities in between. Otherwise, we are going to too much of the effort and then do not get the return. Yes, there's a lot of similarities, but a lot
Okay. Thank you. And my second question is we know that obviously the AI demand and events note, events, leading age, packaging is very tight. But I'm just wondering that in the street wide, we still see probably over capacity in matured note. Yeah, TSMC also have like more mature 16 or nanometer or above that kind of process. So can we kind of a causality?
So Laura is asking how uh fungible are these different packaging Technologies, how interchangeable or easy to transfer the technology between different packaging Solutions. Of course, there are some similarities in between otherwise, though we are going, uh, to take too much of the effort and then did not get the return. Yes, yes. A lot of similarities, but a lot of varieties also,
Our mature nose to kind of make better efficiency and probabilities to enhance the...
Speaker Change: Okay, thank you. And my second question is, uh, we know that obviously the AI demand Advanced node Advanced, uh, Leading Age. Uh, packaging is very tight and um, but I'm just wondering that industry wise. We still see probably over capacity in mature node. Your tsmc also have like, uh, um, more mature, 16 or or, um, nanometer or more above that kind of process. So can we kind of consolidate
Those capacity to fulfill the demands across the board.
Okay, so no second question is on mature notes. She notes there there's over capacity on the industry wide in older notes so she wants to understand for TSMT specifically if we take, for example, 16 nanometer and older notes. What is our strategy? Can we consolidate amongst the different notes? How do we protect our profitability? I don't think so.
Uh, our mature nose to kind of, um, make better efficiency and profitability to enhance the like uh, those capacity to fulfill the demands across the board.
Good question. If you read the newspaper, there's so much of a...
Speaker Change: Specifically, if we take, for example, 16, nanometer and older nodes. Uh, what is our strategy? Can we consolidate, uh, amongst the different nodes? How do we protect our profitability?
Let your know the capacity of Taiwan Semiconductor Manufacturing Co. Ltd.
Speaker Change: Good question. Uh, if you read a newspaper, there are so much of
TSMC's strategy actually is on the mature-known technologies we developed, kind of special.
for example
Speaker Change: Uh, mature node capacity but tsmc's strategy actually is on the mature node Technologies. We develop a kind of specialties
Technology or CMOS sensor or the high voltage
uh, for example that are uh,
Speaker Change: Technology or simos me sensor or the high voltage.
We develop the technology at the request of our customers
Speaker Change: So,
So, we don't worry too much about what you say.
Speaker Change: We develop the technology as a request. So our our customer.
Overcapacity, if it is really overcapacity, we will not build a fab in Japan, we will not build a fab in Germany.
So it's not over capacity. It's all related to customer's need, customer's demand, and those are all specialty technologies.
Speaker Change: So we don't worry too much about what you say the uh, over capacity if it is, really over capacity, we will not build a Fab in Japan. We will not build a fabric in Germany.
Thank you, Laura. I think in the interest of time we'll go to Brad Lin from Bank of America, then we'll take one more from the line and then if there's one more from the floor.
Speaker Change: So it's not over capacity, it's all related to customers that need customers that demand and those are all specialty Technologies.
Speaker Change: Did I answer your question? Thank you, thank you. Okay, thank you. Laura. I think, in the interest of time, we'll go to uh, bradlyn from uh, Bank of America. Then we'll take 1 more from the line. And then if there's 1 more from the floor,
Thank you for taking my question after two questions. My first one is on the human robot. So we have learned that human robot started to contribute to TSMC and it is getting more mentioned as the next frontier of the AI hardware. How does TSMC evaluate the market size of human robot in the semiconductor and in terms of the potential marketamp compute and also sensor requirements?
and with, do you think that might be another driver potentially for mature notes to do?
Okay, thank you, Brad. So, Brad's first question is around human-noid robots. He was starting to see some contribution. He wants to understand, how do we evaluate the market size? What is the addressable opportunities for TSMC in the long-term at the leading edge and also on the mature notes with certain type of specials?
Uh, thank you for taking my question. I have the 2 questions that my first 1 is on the human robot. So we have learned that uh, you know, human robot started to contribute to tsmc and it is getting momentum as the next Frontier of the AI Hardware. How does tsmc evaluate the, uh, Mackie size of, um, human robot in the semiconductor? And in terms of the potential Market Tam compute. And also, since the requirements thank you. And with do you think that might be another driver potentially for mature Notes too? Thank you. Okay, thank you. But so Brad's first question is around, humanoid robots. Uh, he was starting to see some contribution. He he wants to understand. How do we evaluate the market size? What is the uh addressable opportunities for tsmc uh, in the long term uh at the Leading Edge and also on the mature notes with certain type of specialty,
Brett is too early, actually, too early to say the human noise robot will appear raw in this year. Next year, probably still too early because it's so complicated. You know that the human noise robot will be, most of the time, will be used. I think the first one will be used again.
the technical care of the people getting all like me.
Speaker Change: Uh, price is too early. Actually it's too early to say the humanoid robot or play a role in this year. Next year. Probably still too early because of it's so complicated. You know, that the human noise, a robot will be most of the time will be used. I think the first 1 will be used in uh, in medical industry.
You know, I probably some day later I need some human knowledge.
Speaker Change: to take care of the people, getting all like me and uh,
to help me. But...
You know, it's very complicated because we are talking about the brand only. Actually, we are talking about a lot of sensor, sensor technology, that the image sensor, the pressure sensor, the temperature sensor, and all the feedback to the CPU. And so, it's very complicated, and since it's dealing with human being directly, has to be very, very careful.
Speaker Change: you know, I, uh, probably someday that I need some human know, it's, uh, Road robot to help me, but
Speaker Change: You know, it's very complicated because in that we are talking about the brain on the actually talking about a lot of sensor.
But then once you start to fly, it was a big, big blast. I talked to one of my customers.
Speaker Change: Sensor technology that the image sensor, the pressure sensor, the temperature sensor and all the feedback to uh the CPU. And so it's very complicated and since is dealing with uh uh human being directory, has to be very, very careful.
and he said that the EVK is nothing, his robot will be ten times old.
I will wait you for that.
But then once you start to fly, you are a big big plus. I talked to 1 of my customer and he said that the EV car is nothing. He's uh he's a robot or be 10 times of that.
Okay, is that your question? Yes, yes, I believe the client definitely owns EV cars and robots too, so he knows he will So my second question will be on the potential point in a head of the so-called recycling value into 2026 So we know, well normally we continue to recycle the value into our pricing So given the potentially higher pricing into 2026 Are you observing any signs of a demand point from the customers in the second half of the year and potentially well given the tight pipeline of N3 and 5 which we see a continued strength into focus even though we already guided potential decline but yeah, any point
Speaker Change: I'm waiting for that.
Okay. Does that answer your question? Uh yes. Yes I believe that the client definitely owns EV cars and robots too so he knows me. Well,
Okay, Brad's second question very specifically, he's asking as we see he talked about that we will continue to earn our value, do we see any customers trying to pull in their demand ahead of 2026 into the second half of this year, and do we have any additional comments
Well, the answer is no. We did not see any different customers of behavior so far. Okay. But let me show you with you a little more color.
Speaker Change: We see any customers trying to pull in their demand ahead of 2026 uh into the second half of this year and do we have any additional comments to offer on the fourth quarter? Besides what we have already shared?
No, we did not see any uh different customers of behavior so far.
If you are talking about the three nanometers demand, for example,
Speaker Change: Okay, but let me share with you, I add more color.
In the cycle time issue, what takes about the four months?
Speaker Change: Um if you are talking about the 3 Nano meters uh demand for example.
So just no way you can pour in any <expletive> .
Speaker Change: In the cycle. Time is still what takes about the 4 months?
I mean, that's a, yeah, and we have a, as I said, our capacity is very, very tight.
Speaker Change: So, just no way you can pull in anything.
So we already have all the schedule.
Speaker Change: I mean, that's a yeah and we have a, as I said, our capacity is a very, very tight.
Speaker Change: So, um, we already have all the schedules.
Brett Dino's room for poing. Let me show you.
Speaker Change: and so,
Even they wanted, but no, the answer is no [inaudible]
So 2026 is 2026, we will show.
3 little room for pooling. Let me say that even they wanted it. But no, yeah, so it's no.
Thank you, CC, thank you, Brad. Alright, our operator, let's go to the, we'll take questions from the last participant on the line.
Speaker Change: So 2026 is 2026. We will share with you. Yeah, thank you.
Speaker Change: Thank you CC. Thank you Brad. All right. Uh operator. Let's go to the, we'll take questions from the last participant on the line.
Yes, the last one to ask questions, Mati Hussaini from Khir Sai Ji.
Speaker Change: Yes. Oh the last 1 to ask question.
Thank you for taking my question. The first one has to deal with capital intensity. When I look at the past five years, when you were ramping entry and entry, the capital intensity was at about 40%, and you also highlighted how entry and take-out is tracking better than entry and entry combined. Does that imply that the capital intensity would need to go back up to 40%, and in other words, an initial investment for entry to accommodate these take-outs is that the arrival of thinking about how investments are going to play out and have fun.
Me husseini from Sig.
Speaker Change: Yes, sir. Thanks for taking my question. The first 1 has to do with capital intensity when I look at the past 5 years, when you were ramping N3 and N5,
We couldn't hear you exactly clearly. Let me try to summarize your question, alright? Which is, these questions are on capital intensity. He notes that in the past, when we invest in new nodes or structural omega trends, like we have with N38 and 5, our capital intensity has jumped up to greater than 40%. So if I heard you correctly, but your question is this time we talked about the strong demand for 2 nanometer, multi-year upcoming, what is our expectation on capital intensity? Is that correct, matey?
Speaker Change: The capital intensity was at or about 40% and you also highlighted how into PayPal is tracking uh, better than M3 and M5 combined does that imply that the capital intensity would need to go back up to 40%. And in other words, um an initial investment for N2 to accommodate these pay outs is that the level of thinking about how Investments are going to play out and have fun.
Uh, okay, me uh sorry we couldn't hear you. Exactly. Because they let me try to summarize your question. All right. Uh, which is his mi's questions around Capital intensity? He notes that in the past when we uh, invest in in new nodes or or structural the mega trends, like we have with N3 and N5. Our Capital intensity has jumped up to greater than 40%. So, if I heard you correctly, me your question. Is this time? We talked about the strong demand for 2. Nanometer multi-year. Uh, uh upcoming, uh, what is our expectation?
On Capital intensity. Is that correct me?
Yeah, that's correct. Okay. Okay, let me answer this question.
Speaker Change: Yes, that's correct. Okay.
We just said the capital expenditure invested in any year is the future growth opportunities. So if we do our job right, the growth in the next few years is likely to exceed the growth in KPACs dollars, even though, as I said, KPACs dollars is unlikely to drop significantly in the year. So you see a higher growth in revenue than the growth in capital expenditure, then you don't have it such a high capital intensity. We actually demonstrated that in the past few years. And also, let me just share with you that because of this, we're not operating setting a capital intensity as a goal. The dollar among invested is really on the structural demand growth in the following years. So talking about capital intensity is also less meaningful than before.
Speaker Change: Okay, me, let me answer this question.
Speaker Change: Uh, as we just said, the capital capital expenditure, uh,
Speaker Change: Invested in any year is the future growth opportunities. So, if we do our job, right? Uh, the growth in the next few years uh is likely to exceed uh the growth in kex dollars even though uh as I said the kex dollars is unlikely to drop significantly in every given year. So you see a higher growth in Revenue uh than the growth in capital expenditure then you don't have it such a high Capital intensity. We actually demonstrate that that uh in the past few years uh and also let me just uh share with you that because of this uh we're not operating setting a capital intensity as a goal. It's uh the dollar amount invested uh is really on the structure demand growth in the following years. So talking about Capital intensity is also uh less meaningful.
Speaker Change: Than before.
Thank you. Thank you. And a follow-up for me, you highlighted N2P, I'm sorry, you highlighted N16 which will be very applicable for high-possumous compute. Is that the node with AI and HPC with actually a call with a smartphone as an M market that would drive demand for the most
Speaker Change: Thank you. Okay thank you and a follow up for me. You highlighted. Uh m2p uh I'm I'm sorry. You have an M16 uh which will be very applicable for high performance compute.
Speaker Change: Is that is that the note where?
Ai and HPC would actually be at Power with a smartphone as an End Market, that would drive demand for the most illuminated note.
Sorry, Maddie, again, I apologize. We apologize. I could not hear you clearly, but I think his question is about on A16 where we said it's more for a specific HPC related offering. So his question is, I think, Maddie, your question is, is that where the AI demand also comes in for the two nanometer family?
Speaker Change: Sorry me again, I apologize. We apologize. I could not hear you clearly, but I think his question is about, uh, on A6 where we said, it's more for a specific HPC related offering. So, his question is, I think, Mandy, your question is, is that where the AI, uh, demand also comes in for the 2? Nanometer family.
and an asset because so far AI has been N plus 1, N plus 2. Is that node N16 and the first node where AI would move to the leading edge?
Yes.
Speaker Change: I don't know answer because so far AI has been n. Plus 1, n plus 2.
Is that node n62 and the first node where AI uh would move to the Leading Edge?
Okay, maybe let me rephrase it, I think I understand better. His question is really about AI adoption of the leading edge node, the end node. You know, we see a smartphone, we see HPC. His question is very specifically, how do we see the AI adoption of the most leading node for TSMC? He observes in the past, it has generally been one node behind, so how do we see that going forward with things such as ACE?
Well, Maddie, you are right. Thank you, Jeff.
Speaker Change: Okay, his question. Okay. Maybe let me uh uh rephrase it, I think I understand but his question is really about uh AI adoption of the Leading Edge node, the end node. Um, you know, we I see smartphone, we see HPC his question, very specifically, how do we see the AI adoption of the most leading node for tsmc? He observes in the past. It is generally been, uh, uh, 1 uh, node behind. So how do we see that going forward with things such as A6?
Usually, the SPC is a customer, always...
Well me you are right. Thank you, Jeff.
One step behind using M-plus-1 or M-plus-2 technology [inaudible]
Now, because of AI demand, it's so strong, that's one thing but the most important thing.
Usually the SPC is a customer or with a 1 step behind the using M, plus 1, or M, plus 2 Technologies.
You need some kind of performance, but the power consumption is very, very important.
And when we talk about the age of 16, we have another power efficiency improvement across to 20 percent. That's a big value for all the AI data centers, PK.
Speaker Change: Now because of AI demand is so strong, that's 1 thing. But the most important thing is you need some kind of performance. But the power consumption is very, very important.
So, that-
And when we talk about a 16, we have another uh, Power efficiency Improvement across to 20%. That's a big value for all the AI uh, data centers applications.
to help my customer moving faster because of every time when we talk about the AI data center, if you notice that the first thing they talk about is a power supply electricity, right? And TSMC is a technology by the way, and A16 is a further improvement of the N2 node, so it's not a surprise for TSMC to expect for those people in the AI data centers industry they want to use in Asia.
Speaker Change: so that
Speaker Change: uh, help my customer moving faster because of, uh, every time when we talk about the AI data center, if you notice that the first thing they talk about is a power supply electricity, right? So they did not tell you say, uh, the power efficiency is very important, but they tell you that we are to build a very big electricity power plant, to support the AI data center. So that tell you how important it is and tsmc is the technology by the way. Uh, on a 16 is a further Improvement of the N2 node. So, it's not a surprise for tsmc to expect, uh, for those people, uh, in, uh, AI data centers, uh, industry. They want to use in a 16.
Okay. Thank you, CC. Thank you, Midi. We'll take the last question from before. We have one participant here, Felix Pan from KGi.
Okay, thank you CeCe. Thank you. Mighty will take the last question from the floor. We have 1 participant here. Uh, Felix pan from KGI.
Thanks, Jeff. Gav Nino CC and Wendell, I only have one question about the overseas expansion. I think CC earlier mentioned that the second fact for the N3, there's a strong demand, so you guys need to speed up several quarters for that. Together with I think the US government also raised the investment tax credit cap for next year. So I wonder how this shape or how this speed up your ramping schedule for the second fact and how this impact to the overseas fat dilution for the guidance windows given earlier.
Felix Pan: Hi. Um,
I think the follow-up question will be, if you guys speed up the US investment, how is that impact to other regional investment like the Japan and Germany as well? And lastly, is that possible to break down the overseas capex and domestic capex going forward? Yeah, that's all my question thing. Okay, that's pretty much two questions, but okay, so Felix's question on our overseas expansion plans, he notes that yes, he said we're speeding up the schedule for the second fact in the US. So how, and he also notes the recent passage of the US ITC bill. So how does this impact or effect our rent schedules in our US expansion? And what is the implication or impact to the overseas dilution? That's number one.
Uh, thanks Jeff, um, good afternoon CC and window. Um, I only have 1 question about the uh, the overseas expansion. I think CC earlier mentioned that the, the second Fab for the N3. Uh, there's a strong demand, so you guys need to speed up several quarters for that together with. I think US Government also raised the, uh, the investment tax credit, uh, cap, uh, for next year. So, uh, I wonder how this shape or how this speed up your ramping schedule for the, uh, this the the second Fab and how this impact to the overseas, uh, Fab dilution for, uh, the the, the guidance, uh, Windows, uh, give it earlier. Okay? Yeah, sorry. Yeah. And, and I think, uh, the follow-up question will be, uh, if you guys speed up the US investment, how how you is that impact to other uh Regional uh investment like the Japan and Germany as well. And lastly is that possible to break down the overseas capacity?
Felix Pan: Texan.
That's two questions.
Speaker Change: Domestic capex. Going forward. Yeah, that's all my questions. Thank you. Okay. That's, yeah, that's pretty much, uh, 2 questions. But, uh, okay. So but uh, Felix question on our overseas expansion plans, he notes that, yes, CeCe said, we're speeding up the uh uh schedule for the second 5 in the US. Um, so how and he also notes, the recent passage of the US, uh, ITC uh, bill. Um, so how does this impact or or, uh, affect our ramp Pro schedules, in our us expansion and what is the implication or impact to the overseas dilution? That's number 1.
Felix Pan: Wow, that's 2 questions.
Okay, let me share with you about our rainbow schedule. It's totally because of our customers
okay, let me, uh,
and we appreciate the U.S. government increase the ITC from 25% to 35%. We appreciate that. It helps. But the real schedule is because of our customers at Taiwan. So we have to prepare the capacity to meet your at Taiwan. That's a number one consideration.
Felix Pan: Share with you about the our rebar schedule is totally because of a customer's demand.
Felix Pan: And we appreciate the US government, increase the ITC from 25% to 35%, we appreciate that it helped.
But the real schedule is because of our customers demand. So we have 2 prepare the capacity to Media demand that's the number 1 consideration.
Ma Zuyao
The margin impact, it is positive, although not that significant in the five year period. Think about this. The ITC is used to offset the Yasm.
and the benefit comes when depreciation starts. So it gets M.
Felix Pan: The margin impact, it is positive. Uh, although not that significant in the 5 year period, think about this, the ITC is used to offset the asset value and the benefit comes uh when depreciation starts. So it gets advertised
And then Felix's second question is how does the ramp and speed up of the U.S. Expand cluster expansion. How is this impacting our expansion plans in Japan and Europe ?
Well, you think about the TSMC expansion, the over-C5. In the US, it's a leading edge. In Japan is a special technology to be specific. Most of the time, it's for the CMOS emissions.
Felix Pan: And then Felix is. Second question. Uh, is, how does the ramp and speed up of the US? Uh, extension cluster expansion? How is this impacting, our extension plans in Japan and Europe?
for Germany is automotive industry.
Speaker Change: If it does at all. Well, you think about the tsmc expansion. Uh, the oversea 5 in the US is the Leading Edge in. Japan is uh unspecial technology uh, to be specific. Most of the time is of for the simos emission.
So, they are all not in the shampoo. So, actually, it's not affect the investment in the U.S. or
Speaker Change: uh, for Germany is um, automotive industry
Thank you, CC. Thank you, Felix. Okay, everyone, so this concludes our question and answer session. Before we conclude today's conference, please be advised that the replay of the conference will be accessible within 30 minutes from now. The transcript will become available 24 hours from now, and both are going to be available through TSMC's website at www.tsmt.com. So thank you very much for joining us today. We hope everyone continues to stay well, and we hope you will join us again next quarter. Goodbye. And have a good day. Thank you.
Speaker Change: So they are all not in the same field. Also, actually, it's not affect the investment in the US or investing on the Leading Edge does not affect the investment in Japan or in Germany.
Speaker Change: Thank you CeCe. Thank you Felix. Okay, everyone. So this concludes our question and answer session, uh, before we conclude today's conference, please be advised that the replay of the conference will be accessible within 30 minutes from now, the transcript will become available 24 hours from now and both are going to be available through tsmc's website at www.psn.com. So thank you very much for joining us today. We hope everyone continues to stay well uh and we hope you will join us again next quarter. Goodbye and have a good day. Thank you.