Q4 2024 Waterdrop Inc Earnings Call
Oh.
Yeah.
Yeah.
Yeah.
Yeah.
Okay.
Tracy Li: Good morning, everyone. This is Tracy Li from Waterdrop Investor Relations.
Speaker Change: Good morning, everyone. This is Tracy Li from Investor Relations.
Tracy Li: It's my pleasure to welcome everyone to Waterdrop's fourth quarter and official year 2024 earnest conference call. All participants are in listening only mode in our English line. As a reminder, today's conference call is being recorded.
Speaker Change: It's my pleasure to welcome everyone to when he got fourth quarter artificial ear company for earnings Conference call. All participants are in listen only mode in our English line. As a reminder, today's conference call is being recorded. Please note that discussion today will contain forward looking statements made and that's the copper position you are secured.
Tracy Li: Please note that discussion today will contain forward-looking statements made on the safe harbor provisions of U.S. private securities and the Litigation Reform Act of 1795. Forward-looking statements are subject to risk and uncertainties and may cause actual results to differ maturely from our current expectations. Potential risk and uncertainties include but are not limited to those outlined in our public findings with the SEC. The company does not undertake any obligation to update any forward-looking statements except as required under applicable law.
Speaker Change: And the litigation Reform Act up 1995 forward looking statements are subject to risks and uncertainties actual results could differ materially from our current expectation potential risks and uncertainties.
Speaker Change: Turning his include but not limited to those outlined where public filings with the SEC.
Speaker Change: Company does not undertake any obligation to update any forward looking statements.
Speaker Change: As required under applicable law also all these call include discussion of certain non got here.
Unknown Executive: Also, this call includes discussion of certain Please refer to our earnings release for more confirmation between NGAP and GAP. joining us today.
Speaker Change: Please refer to our earnings release for more of a combination between non-GAAP and GAAP.
Speaker Change: Joining us today.
Tracy Li: On the call, Mr. Chen Peng, our founder, chairman, and CEO. Mr. Ren Wei, Director and GM of Insurance Business. Xu, Head of Finance Department, and Ms. Jasmine Li, Board Secretary.
Speaker Change: On the call Mr. Shin Kong, our founder Chairman and CEO.
Mark: Mr runway character and GM Mark insurance.
Speaker Change: And then she has got one.
Speaker Change: Finance Department and I missed it.
Speaker Change: Definitely in the fourth century will be happy to take some other questions in the mineral light at the end of the conference call.
Unknown Executive: We'll be happy to take some of the questions in the mentoring line at the end of the conference call.
Peng Shen: Hello, everyone. This is Peng Shen. In 2024, Waterdrop delivers solid performance across all business units, driven by our integrated scenario ecosystem and the sustainable AI investment. By fully upgrading products and services, we have established new benchmarks for AI implementation in the industry. In Q4, our total revenue was about $687 million. The annual revenue reached $2.77 billion, up by 5.4% worldwide. In terms of the profits and net profits attributable to ordinary shareholders, in this quarter, it was over $99.6 million, and the annual profit result has doubled, achieving our financial guidance to the market. The company has remained profitable for 12 consecutive quarters.
Speaker Change: Hello, everyone. This at some point in 2024, what is the op deliver solid performance across all business.
Speaker Change: Driven by our inter.
Speaker Change: Integrated arrow ecosystem and it's sustainable.
Speaker Change: I fully upgrading credit and services, we have establish new benchmark for AI implementation in the industry in Q4. Our total revenue was about 687 million then the annual rate of revenue reached $2 77, giving back up by 5.4% Y O Y in terms.
Speaker Change: Perfect.
Speaker Change: Actually if you go to ordinary shareholders in the quarter with over 99.6 milligram and the annual appropriate Brent also had doubled and achieving our financial guidance to the market the comedy or has remained profitable for 12 consecutive quarters.
Peng Shen: I'm also pleased to announce that based on our stable operation in the second half of 2024 and a significant year-over-year increase in profits, with board approval, the company will soon launch It's the third cash dividend, with 0.2 cents per ordinary shares and 2 cents per ADS. The shareholders of ordinary shares and ADS as of the closing date of April 11, 2025 U.S. Standard Time are entitled to the dividend. Moreover, if our business maintains healthy operations, we plan to distribute dividends semi-annually to reward the support of our shareholders and investors.
Speaker Change: I'm also pleased though now that they've gone were stable accretion in the second half on 24.
Significant year over year increase in profits with board approval the company with the one launched.
Speaker Change: It's the third tested it.
Speaker Change: The cash dividend when theyre going to spend per ordinary shares and to answer.
Speaker Change: The shareholders ordinary shares and ADR or clothing data April 11, 25 U S. Send your time are entitled to the dividend well over an hour.
Speaker Change: Mackenzie healthy operation, we plan to distribute dividends semi annually.
Speaker Change: Toward the support of our shareholders and investors.
Peng Shen: In addition to dividends, the company also continues to actively repurchase from the open market. As of the end of February 2025, the company has cumulatively repurchased about $52.1 million ADS in the open market, costing approximately $104 million US dollars.
Speaker Change: In addition to dividends our the comedy also continue to actively repurchase from the open market.
February 25, the company has cumulatively repurchased about $52 1 million in open market costing approximately 100 and inform you in U S dollars.
Peng Shen: We remain committed to social responsibility and sustainable development. As of the end of 2024, Waterdrop Welfare Platform has partnered with 115 public charities and hosts over 15,000 charitable projects. Furthermore, we also won many ESG achievements in Q4, including AESG rating from United, Equator, and Trudy. As a milestone, Waterdrop Medical Co-founding has been officially recognized as a pioneer in China's medical co-founding sector. Following new legislation changed, the platform secured designation from the Ministry of Civil Affairs on December 25, 2024, becoming one of the first batch of state-endorsed online service platforms for individuals seeking financial help.
Speaker Change: We remain committed to social responsibility and sustainable development as of the end.
Speaker Change: So one way for what is a welfare platform.
<unk> has partnered with hydro <unk> colleague charities and hope to over 15000 charitable projects. Furthermore, we will also won many ESG achievements.
Speaker Change: Yeah in Q4, including a yes your ratings Earl.
Speaker Change: I view that equator and intriguing.
Speaker Change: As a milestone in what is a medical for bonding happening officially recognized as a pioneer in China medical clarifying factor.
And you like less switched and changed the platform securely violation from the Ministry of Civil Affairs on December 25 to $1 four becoming one of the first batch of Cid. It endures online service platform for individuals seeking financial help.
Peng Shen: Since China's Spring Festival, DeepSpace has sparked global attention in AI innovation. We believe AI-powered industry will redefine productivity. As a tech-driven company, we have launched AI tools and agents using large-language models in 2024. especially in online sales and broker support. With DeepSeat integration, we are accelerating AITC development.
Speaker Change: Chinese Spring Festival deep space has sparked.
Speaker Change: Global attention in AI innovation, we believe AI powered English Street.
Speaker Change: <unk> redefined productivity as a tax driven company, we have launched AI tools.
Speaker Change: Using the last leg remodels in 'twenty four.
Speaker Change: Especially in online itself.
Speaker Change: Broker support with deep integration, we will.
Speaker Change: Our star rating ICT development at the end of today's earnings call, we will share our insight and the latest breakthrough.
Peng Shen: At the end of today's call, we will share our insight and latest breakthrough on AI. For 2025, we will focus on business growth, targeting 20% annual revenue growth with quarterly increments. Based on current projections, we expect financial performance of Q1 to maintain stable growth and sustain profitability. Subsequent quarters may see strategic investments aligned with federated revenue growth.
Speaker Change: AI.
Speaker Change: Well, it's on Wifi, we will focus on business growth.
Speaker Change: 20% annual revenue growth with quarterly increments based on current projections, we expect financial performance of Q1 Q2.
Speaker Change: Stable growth and sustained profitability.
Speaker Change: Subsequent quarters, Macy's strategic investments aligned with fabricated revenue growth.
Peng Shen: That concludes my remarks.
Speaker Change: That concludes my remarks, now I will turn it over to Sharon for the financial update.
Xiaoying Xu: Now I will turn it over to Xiaoying Xu for the financial update. Thank you, Shen Peng. Before I go into details, please be reminded that all numbers quoted here will be in R&B and please refer to our earliest release for detailed information on our financial performance on both the year-over-year and quarter-over-quarter based risk factors.
Sharon: Thank you Paul.
Speaker Change: Before I go into detail please be reminded that all numbers quarter here will be RMB and please refer to our earnings release for detailed information on our financial performance on both the year over year any carnival auto being fifth respectively.
Xiaoying Xu: In Q4, the company delivered a solid performance with total revenue to reach $687 billion. at 4.2% of YOY for a full year 2024. Our revenue total is $2.77 billion. In 2024, our non-insurance business accounted for near 15% of total revenue. Our crowdfunding service fees amounted to $267.7 million, while digital clinical trial solution income reached $91.1 million, collectively reflecting diversified revenue stream. Through AI-driven efficiency enhancement and refined operation management, our total operating costs and expenses in Q4 amounted to about $634 million, remaining stable compared to the same period last year. For the full year of 2024, overall operating costs and expenses decreased by 1.7% from last year.
Speaker Change: In in Q4, the company delivered a solid performance with total revenues reached 687 billion.
Speaker Change: At four 2% on a Y O Y four for full year 'twenty, one for our revenue totaled $2 77 billion.
Speaker Change: In 24, or 90 insurance accounting for near 15%.
Speaker Change: So our revenue.
Speaker Change: Our corresponding service fees amounted to 267 7 million.
Speaker Change: Stevensville claim could fly associate income reached $91 1 million collectively reflecting diversified revenue stream.
Speaker Change: Through AI driven efficiency.
Speaker Change: It has been and refined operating spend management, our total operating cost and expenses in Q4, an issue about 634 million remaining stable compared to the same period last year.
Speaker Change: The full year of 'twenty, 'twenty, four or all operating costs and expenses decreased by one 7% from last year.
Xiaoying Xu: Breaking this down, in Q4, operating costs increased by 7.2% year-over-year to $323.8 million. This was primarily driven by an increase of $39.1 million in cost of referral and service fees, partially offset by a decrease in personal costs. sales and marketing expenses increased by 4.1% year-over-year to $182 million this quarter. This resulted from our strategic expansion in customer acquisition due to an increase of $30.2 million in marketing expenses for third-party traffic channels. and partially offset by a decrease of $22.8 million in personal costs. General and administrative expenses declined significantly to $73.7 million, down 23.2% year-over-year, nearly attributable to a decrease in allowances for credit loss and professional services.
Speaker Change: Breaking this down in Q4 operating costs increased by seven 2% year over year to 323 8 million.
Speaker Change: This was primarily driven by an increase.
Speaker Change: $39 1 million from referral and Turkey, partially offset by a decrease in personal cost.
Speaker Change: Sales and marketing expense increased by four 1% year over year to 182 million in the quarter that resulted from our strategic expansion in customer acquisition due to an increase.
Speaker Change: 32 million in marketing expenses for certain third.
Speaker Change: Third party traffic channel.
Speaker Change: And partially offset.
Speaker Change: All set by a decrease of $22 8 million in personnel costs.
Speaker Change: General and administrative expenses declined significantly to $73 7 million down 23, 2% year over year, mainly attributable to a decrease in allowance for credit loss.
Speaker Change: Professional service fees.
Xiaoying Xu: R&D expenses stood at $54.3 million, a 9.3% decrease from the same quarter of last year, primarily reflecting optimized fiscal costs. However, our AI investment remains unwinding. In fact, we plan to increase investments in 2025 to further strengthen our technology edge. We have sustained remarkable profitability and double our annual net profit attributable to our ordinary shareholders. We continue to generate a positive operating cash flow during this quarter. As of December 31, 2024, our position was $3.67 billion, providing a robust support for future growth initiatives.
Speaker Change: R&D expense totaled at $54 3 million again.
Speaker Change: 3% decrease from the same quarter, a pop here, primarily reflecting optimized this personal cost.
Speaker Change: However, our AI investment remains on Whiting.
Speaker Change: And in fact, we would plan to increase investments to further strengthen our technology edge.
Speaker Change: We have sustained a remarkable.
Speaker Change: Profitability and double our annual net profit attributable to slower ordinary shareholders. We continue to generate a policy of operating cash flow during this week.
Adam: Thats correct Adam.
Adam: 31 2024.
Adam: Position with three points 67 billion.
Adam: Providing a robot the support for future growth initiatives.
Xiaoying Xu: And this concludes our financial overview.
Adam: And this concludes our financial overview I will now pass to our Board Secretary Rich asked me to update our business development.
Jasmine Li: I will now pass to our Board Secretary, Jasmine, to update our business development. Thank you, Xiaoying. In Q4, our insurance business achieved a FIP of about $1.9 billion, representing 24.5% year-over-year growth, through showing a 7% quarter-on-quarter decrease. Insurance-related revenue reached $580 million, down 1% YOY and 3% quarter-on-quarter, while operating profits stood at $103 million with an 18% operating margin. During this quarter, our short-term insurance FIP surged 44.2% worldwide to 1.43 billion yuan. Driven by breakthroughs in public customer acquisition through the real-time marketing optimization and the new payment channels leveraging user insights, we introduced more inclusive short-term insurance products in this quarter, including medical plans tailored for individuals with pre-existing conditions.
Speaker Change: Thank you Dr Yang and in Q4, our insurance business achieved.
Adam: About $1 9 billion.
Adam: Representing 24.5% year over year growth sterile showing.
Adam: 7% in the quarter on quarter decrease insurance related revenue reached 580 million is down 1% Y O y and 3% quarter over quarter, while operating profit stood at 103 million with an 18%.
Adam: Operating margin.
Adam: During this quarter, our short term insurance up IP search going for 2% Y O Y to 143 billion driven.
Adam: Driven by breakthroughs in public customer activation through the real time marketing optimization, and the new payment channels leveraging user insights.
Adam: Introduce more inclusive short term insurance product in this quarter, including medical clients tailored for individuals with preexisting conditions.
Jasmine Li: For long-term insurance products, FYP reached $460 million with a slight sequentially decreased impact by the product suspension of last quarter and the holiday rhythm in this quarter. In Q4, we optimized our team staffing while maintaining high productivity per capita. Furthermore, we sustained our efforts to enhance and diversify our long-term insurance offerings with disability income insurance, along with achieving an FYP existing in 90 million. The company is also continuing to develop its online planner model. In the fourth quarter, Shen Lanbao independently contributed approximately 90 million yuan in FIQ. For our medical performing business, as of end of 2024, around 470 million people accumulated.
For long term insurance.
Adam: <unk> reached 406.
Adam: 16, New England with.
Adam: Slide sequentially decrease impacted by the suspension of last quarter and the Halloween event in this quarter.
Adam: In Q4, we optimized our team Duffy well maintaining a high productivity per capita.
Adam: Furthermore, with sustain our efforts to enhance and diversify our longtime insurance offering with <unk>.
Adam: Ability income insurance alone achieving and.
Adam: Existing 1990.
Median.
Adam: The company is also continue to develop its online planner model in the fourth quarter Shannon ball independently independently contribute approximately 90 million in.
Adam: In FY <unk>.
Adam: For our medical profiling.
Adam: I heard before around the 417 million people accumulate ABB. They donate had $67 5 billion to about $3 4 million pension Cisco our platform our platform.
Jasmine Li: They donated $67.5 billion to about 3.4 million patients through our platform. Our platform processed thousands of campaign applications and over 50,000 images. image submissions, with risk control remaining a top priority. To enhance efficiency, we implemented a dual-model AI framework. On one hand, lightweight models now detect pixel-level image tampering in milliseconds, reducing false positive rates from 0.3% to in the three leading 0.1%. On the other hand, summarizing the experience from the accumulated cases from the past eight years, extracting more than 1,700 characteristics to assist medical consultants in making reasonable prediction of the diseases and reducing the risk of cause by fraud and misappropriation.
Adam: Chris asked a thousand of companion applications and over 50000 image.
Adam: You made a submission with risk control remaining a top priority to enhance efficiency we implemented it.
Adam: <unk>.
Adam: Do model.
Adam: Mark.
On one hand lightweight model now detected pixel level in major tempering in mere seconds.
Adam: Second reducing false positives break from <unk>, 3% to <unk> industry, leading <unk>, 1% on the other hand summarizing.
Adam: Our experience from the accumulated cases from the past eight years.
Adam: Yeah.
Extracting more than one 1700 of characteristics to medical consultants in making reasonable prediction of the diseases and reducing the risk of cop five rod and is the preparation.
Jasmine Li: Following 2024's regulation update and the government's official designations, Waterdrop, as China's largest online service platform for individuals seeking financial help, will continue setting standards for technology-driven transparency and user-centric operations. During this quarter, our digital clinical trial solution income continued its growth, realizing a year-on-year increase of 22.9%. We collaborated with a total of 177 pharmaceuticals and CROs. Moreover, we enrolled 880 patients and initiated services for 79 new programs during the first quarter of 2024. In addition, the platform leveraged AI models to enhance doctor-patient service across scenarios like health management, automatic science content creation, and education.
Adam: <unk> 29 for the regulation of updates and a government official designation what is Rob as China's largest online services platform point of view sticking finished how will continue setting standards. So a technology driven transparency and user centric operation.
Adam: During the quarter, our Janesville clinical trials duration income continue its growth realizing a year over year increase of 22, 9%. We collaborated with a total of 177 pharmaceuticals and zero all over we can grow at 8%.
Adam: Jordan 80 patients and initiatives service for 79 New program.
Adam: During the fourth quarter of 24. In addition, the platform leveraging AI model Doctor patient surveys across narrow like health net has been automatic.
Adam: Science content creation and application.
Jasmine Li: you didn't know all were there. Multiple initiatives have been launched, both in efficiency and user experience, while ensuring compliance.
Adam: <unk>.
Adam: Multiple initiatives have been launched boosting efficiency and user experience well.
Adam: Ensuring compliance.
Jasmine Li: This is a key progress of each business line in this quarter.
Adam: This is a key pro breath of each business line.
Adam: This quarter next we were invited runway the GMO.
Wei Ran: Next, we will invite Li Rang Wei, the GM of the insurance business, to deliver a keynote on Waterdrop's AI-Driven Insurance Hacks. Thank you, dear Ambassador and Analysts. It's a great honor to present Waterdrop's achievements in AI and other intro texts during the earnings conference. Waterdrop has been deploying AI since 2019, starting with machine learning algorithms, OCR, and NLP technology to optimize insurance product recommendations, mismatching, and risk control, boosting online conversion rates and service efficiency. We established a series of digital infrastructure, including a data middle platform, intelligence marketing system, smart CRM system, and AI risk control system.
Adam: Insurance business to deliver a keynote on it wasn't jobs AI driven <unk>.
Jeremy: Thank you, Jeremy Dear Investor and analyst.
Jeremy: It's a great honor to prevent water drops achievements in AI and other intra taxed during the earnings conference.
Jeremy: What does it all have been deploying AI thing.
Jeremy: 2019 are studying with machine.
Jeremy: Machine learning algorithms, all CRE and <unk> and L. P. R Tech.
Jeremy: Knowledge to optimize insurance product recommendation, these matching and risk control boosting online conversion rates and service efficiency.
Jeremy: We established a theory, though digital.
Jeremy: Digital infrastructure, including data platform.
Jeremy: Intelligence marketing system, more CRM system, and AI risk control system enhancing user experience well strengthen our insured has leadership. This journey also cultivated a cross functional R&D team combining insurance expertise with AI capabilities.
Wei Ran: Enhancing user experience while strengthening our intratact leadership. This journey also cultivated a cross-functional R&D team, combining insurance expertise with AI capabilities.
Wei Ran: Underneath this transformative way, Waterdrop has developed Inherit, EnsureTag, competitive barrier for AI applications, which can be divided into four core areas. Firstly, we have built a multi-dimension ecosystem, interacting tags and voice interactions, digital avatars, and other innovations. Covering channels such as WeChat, Wecom, and AI outbounding calls, this enables seamless transactions between services while accumulating multidimensional data. Secondly, we have established a fully digitalized service chain, spanning demand identification, product matching and writing, and claims processing. Unlike the conversational ability of general-purpose LLMs, our service chain requires not only massive knowledge bases, but also precise control over the entire service workflows to ensure high reliability.
Jeremy: And I made this transformation to weigh what is Rob has developed inherit ensure attack competitively barrier for AI applications, which can be divided into four core areas. Firstly, we have huge and multi dimensions ecosystem interacting Tac and the voice interaction digital app.
Jeremy: <unk> and other innovation.
Jeremy: Covering channels, such as <unk> com and <unk> all plumbing call.
Jeremy: This enables seamless transactions between service well.
Jeremy: Well accumulated and multi dimensional if that helps.
Jeremy: Secondly, we have established a poorly.
Jeremy: Fully digital life with every chance bedding demand identification correct matching and are writing and claims with adding Alexa.
Jeremy: So as you know everybody all general purpose Arrow and our service channel requires not only massive knowledge basis, but it also provides control over the entire service workflow to ensure high reliability.
Wei Ran: Currently, we handle tens of thousands of highly complex insurance inquiries in databases. During over 8 years, we process 100 million scale private data assets in the insurance industry, including 112 million operations data on knowledge base of over 7,000 products and massive structured service data entries.
Jeremy: Currently we handle tens of thousands of highly complex insurance in Paris in databases.
Jeremy: Clearly over eight eight years, we <unk> hundred million go private they had at that in the insurance industry, including a 112 million operation data on.
Our knowledge base of over 7000 products and a massive structured as service data entries. This real world data is in repo school training material for building insurance specific L O N.
Wei Ran: This real-world data is irreplaceable training material for building insurance-specific LLMs. Lastly, distinct from general-purpose LLMs, Waterdrop has developed a vertical knowledge graph for insurance scenarios, combining recommendations, insights, and service conversion. Across hundreds of refined operational knowledge, we employ specialized tools such as family coverage gap analysis around customer decision-making, psychology models, and resolution framework for customer objections. For example, our model can generate a customized insurance plan in real time based on user blogging duration, consultation keywords, family structure, and other features.
Jeremy: Luckily it seems.
Jeremy: <unk> from General purpose <unk>, what is the op has developed a vertical knowledge gap.
Jeremy: Breath for insurance this scenario, combining recommendation insight and service conversion.
Jeremy: Across hundreds of very fine I appreciate Illinois, We ink Inc.
Jeremy:
Jeremy: Our specialized towards family coverage gap and ISS, a radar customers design and decision.
Jeremy: Decision, making psychology models and.
Jeremy: Gross resolution framework for customary objection.
Jeremy: For example, our model can generate customer not compromise the insurance client in real time based on either blogging duration.
Jeremy: Consultation keywords family structure and other features.
Wei Ran: is 4-in-1 capability system integrating scenarios, service, data, and knowledge, ensuring the rapid implementation of Waterdrop AI solution. While general purpose LLM still struggle with Haluthi Nations, our vertical models can already deliver definite answers to our users. Overall, the combination about the multiple scenarios and online data has offered irreplaceable strategic value in insurance LLM.
Jeremy: This 14, one capability system integrating scenarios services data and knowledge, ensuring the rapid implementation of what is out there.
Jeremy: Solution well general purpose.
Jeremy: Do you struggle with.
Jeremy: Lucy nation, we our vertical model can already do you remember detonated answers to our users.
Jeremy: Overall with the combination.
Jeremy: The multiple scenarios and online data has offered it and replace go strategic strategic valuable in insurance.
Wei Ran: Next, I will demonstrate how our AI has delivered measurable results. Over the past two years, we have actively adopted cutting-edge fundamental LLM technologies while developing AI applications tailored to our business scenarios.
Jeremy: Next our demonstrated how our AI has to be the burden narrow ago results.
Jeremy: Over the past two years, we have actually adopted cutting age fundamental L. O M technologies, well, it's about helping AI application tailored to our business scenario I want to share some examples.
Wei Ran: I want to share some examples. Firstly, our insurance AI agent has end-to-end self-decision-making capability. It integrates the ability of perception, inference, decision execution. Leveraging Multi-Agent Collaboration Technology In the medical insurance scenario, it enables fully automatic decision making up to 15-minute conversations. For example, a single call can automatically complete a process like understanding user needs, health risk modeling, and the policy interpretation. Our AI agent can support modular agent units collaboration in real time and the flexible transition across health insurance, auto insurance, and other scenarios. The system sprays over 7,000 products and it can respond within two seconds.
Jeremy: For example, firstly, our insurance AI agent half into and south decision, making capability.
<unk> integrated the ability of a recession.
Jeremy: Procession in France.
Jeremy: Decision execution.
Jeremy: Leveraging multi agent collaboration technology in the medical insurance narrow it enables fully automatics.
Jeremy: You said you are making up to 15 minute conversation for example, a single call can automatically calculates pro fast like and there's any user needs how health risk modeling and.
The policy interpretations.
Jeremy: Our AI agent cap report a modular agent units collaboration in real time, and the flexible transaction transition across health insurance.
Jeremy: Insurance and other scenario.
Jeremy: The system right.
Sure.
Jeremy: 7000 products and I can't recall within Tuesday.
Wei Ran: Moreover, our AI agent can dynamically optimize LLM algorithms by learning from the operation feedback and self-outcomes through over 100,000 data interactions.
Jeremy: Moreover, our <unk> AI agent Kent.
Jeremy: <unk> optimized.
Jeremy: Our rhythm by learning from the accretion feedback and South I'll call.
Jeremy: There are over 10.
Jeremy: 100000.
Jeremy: The interaction.
Wei Ran: And secondly, we also continue to optimize our model through two main approaches. The first one is knowledge distillation. This method uses technology like SFT and chain of thought reasoning. It impacts millions of insurance-related knowledge to the model, making it comparable to the senior human computation. The second is the reinforced fine-tuning of the cell strategies. We integrated the DeepSeq and other open-source frameworks with our self-developed GRPO algorithm. This has increased model accuracy by 20%.
Jeremy: And secondly, we also continue to optimize our motto stroke two men approach. The first one is knowledge distillation.
Jeremy: This map technology like S F G and China false resenting it impacts millions of insurance real estate knowledge to the market, making it comparable to the senior human couple of it.
The second is the reinforced fine tuning of the south strategies, we integrate hays the deep seek and other open source frameworks with our self developed loved G. R. P O.
Jeremy: This has increased motto accuracy by 20%.
Wei Ran: Furthermore, our AI agents can make multi-modal interaction with our users, such as real-time coordination of voice, text, and the image, minimizing user waiting time. We also upgraded it with an emotion adaptive system. It can dynamically adjust its speaking speed and modify its tone based on semantic analysis. For example, our AI agent will slow down by 20% when it interacts with elderly users.
Jeremy: The more our AI against can make a multimodal interaction with our users such as real time coordination of boys tax.
Jeremy: And and does the image minimized you they're waiting time.
Jeremy: We also upgraded it with and emotion adaptive system. It can dynamically adjusted speaking speed and the modified PA based on semantic analysis like that who our AIA Jane will slow down by 20% when you interact with elderly users.
Wei Ran: Over the past two years, our 48 LLM-related patents have been granted, with over 100 patents filed and nearly 100 AI innovations. Waterdrop is one of the first insurtech companies capable of enabling AI agents to independently close deals in complex scenarios.
Jeremy: Over the past two years, our 48 yellow and we're gonna they had patents have been granted.
Jeremy: With over 100 patents filed and near the 100 AI innovations.
Jeremy: What is that that is one of the first insurer tax companies capable of enabling.
Jeremy: Oh Boy AI agent to independently close deals in complex scenario.
Wei Ran: Next, I want to show you our detailed AI application results. Following a development path from simple automation to advanced integration, our LLM application now focuses on two directions, co-pilot and AI agent, directing serving users. For our co-pilot application, we have quality control co-pilot and stealth co-pilot. Our quality control co-palette can replace the manual sampling of service recordings and perform text analysis. With 88% accuracy level of human voice across voice and text channels, we expect the front-line quality control cost to be reduced by 40%, significantly improving efficiency. And our sales co-pilot can enhance sales efficiency by enabling rapid summarization of client intent and tailored product recommendations.
Jeremy: Next I want to show you our detailed AI application result.
Jeremy: Following a development path Brown simple automation to advanced integration our L. L. M application now focused on two direction co pilots and AI agent directing serving users.
Jeremy: Oh, our co pilot application, we have quality control pilot and self co pilot.
Jeremy: Our quality control full pallet can replace the manual assembling all surface recording and performed taxing analysis.
Jeremy: With 88% accuracy level, all QM board across voice and text.
Jeremy: Text.
Jeremy: Channels, we expect the frontline quality control cost to be reduced by 40% significantly improving efficiency.
Jeremy: And our south copilot can enhance south efficiency by E ethylene rapid summarization.
Jeremy: Climb.
Jeremy: Clients intend and tailored product recommendation.
Wei Ran: This tool elevates the service quality and conversion rate. empowered our human agent to better meet client needs and achieve self-targets.
Jeremy: This tool elevate the service quality and conversion rate.
Jeremy: Empowered our human agent to better meet client needs and are achieving achieve.
Jeremy: Achieve the targets.
Jeremy: Yeah.
Wei Ran: About our AI agent, we have three different applications, including AI health insurance agent, AI service assistant, and AI insurance planner. For our AI health insurance agents, it can directly serve customers via voice-based conversation. Consultation, like a single session lasting up to 15 minutes, all performing junior agents in a patient. It currently generates millions of insurance premiums per month. project to reach tens of millions by year-end. Building on this, our AI auto insurance agents have entered a daily training with two PNC insurers with selected projects already in user beta testing. or our AI Service Assistant, Bao Xiaohui.
Jeremy: About our AI agent, we have three different applications, including AI health insurance agents AI services systems, and AI insurance planner.
Jeremy: Well, our AI housing insurance agents. It can directly served customers voice based the conversation.
Jeremy: Consultation.
Jeremy: Like things single session lasting up to 15 minutes outperforming junior agents.
Jeremy: <unk>.
Jeremy: It's currently generates millions who have insurance premium per her mountain.
Jeremy: Project to reach tens of millions by year.
Jeremy: Year end.
Jeremy: Beyond this our AI Oh, Tony insurance agents have entered a dented training with two P&C insurers.
Jeremy: Selected projects that are already there better chassis.
Jeremy: Well, our AI service assistance shall we.
Wei Ran: embedded in our app. This 24-7 AI assistant handles Q&A about user policy and the products you are interested in. By 2025, we expect AI to reduce case transferring rate by 50% and improving efficiency by 50% compared to human agents. Leveraging TIPSiX's long-chain reasoning ability, we are developing an AI insurance planner to provide real-time, high-quality family insurance consultation and planning services. enhancing user experience and growth potential.
Jeremy: Embedded in our App.
Jeremy: 24 seven.
Handle Q&A about the policy and the products you are interested in by talking about a five week expect AI to reduce.
Jeremy: Transferring your rates by 50% and improving efficiency by 50% compared to human agents.
Jeremy: Leveraging <unk> long chain of reasoning ability, we are developing an AI insurance plainer to provide real time high quality finally insurance Colocation and planning services.
Jeremy: Enhancing user experience and growth potential.
Wei Ran: and talking about the future outlooks. In the next three to five years, along with the AI technology and the application trends that I just mentioned, based on our industry know-how, we will further consolidate our leading position in the intratech industry.
Jeremy: I'm talking about the future outlooks.
In the next three to five years, along with the AI technology and application trained that can I just mentioned.
Jeremy: Based on our industry know, how we will further consolidate our leading position in the insured hacking industry.
Wei Ran: And thank you for listening.
Jeremy: And thank you for listening.
Unknown Executive: And ladies and gentlemen, with that, we will conclude today's earnings conference call. We do thank you for joining. Have a good time.
Jeremy: And ladies and gentlemen, with that we'll conclude today's earnings.
Jeremy: Earnings Conference call. We do thank you for joining have a good time.
Jeremy: [music].
Jeremy: Yeah.
[noise].
Unknown Executive: and Li Xiaoying Thanks for watching! and others.
Jeremy: Hum.
Jeremy: [music].
Jeremy: Yes.
Jeremy: Yes.
Jeremy: [noise].
Jeremy: Hum.
Jeremy: Hum.
Jeremy: [music].