Ant Afu: Evolving from a Q&A Tool to a Family Health Butler

Overview

This report is a summary of key takeaways from a conference call between Nomura's China Internet team and experts regarding Afu (Ant Afu), a medical AI assistant under Ant Group. Afu is an AI assistant focused on healthcare launched by Ant Group (unlisted, affiliated with Alibaba), providing users with health and medication information and connecting them to hospitals. The report presents expert judgments on Afu's strategic positioning, technical architecture, user data, commercialization path, and competitive landscape, along with Nomura's ratings and target prices for related targets such as JD Health and Alibaba. The core conclusion is that Afu's development focus is on expanding user scale and improving retention; commercialization is still in its early stages, and it is more inclined to complement existing internet healthcare platforms rather than replace them in the short term.

Core views

At the strategic positioning level, experts believe Afu's ambition far exceeds that of a simple medical chatbot. Afu currently consists of three core modules—health Q&A, continuous health management, and medical services: Health Q&A is the most frequently used feature, covering symptom consultation, physical examination report interpretation, medication education, and disease-related questions; Health Management extends usage scenarios to health diaries, weight management, and chronic disease monitoring; the Service layer connects users to hospital registration, online consultations, medication purchasing, and physical examinations. Experts stated that Afu's goal over the next 3-5 years is to develop into a full-cycle AI health butler, covering pre-diagnosis guidance, medical navigation, and post-diagnosis management, ultimately positioning itself as a family health manager rather than a single information tool. On the technical level, Afu uses Ant's Bailin model as the general foundation, supplemented by specialized models in the medical field, and appropriately incorporates other models such as Alibaba's Tongyi Qianwen. Compared to general large models, Afu improves domain accuracy and reduces hallucinations by introducing professional medical data. It has obtained clinical guidelines, medical publications, de-identified consultation records, and examination reports through hospital partnerships for model training and optimization. At the same time, it builds a training flywheel based on real user interaction data under compliant conditions. Regarding user scale and retention, experts estimate that Afu's standalone App DAU is approximately 8-9 million, with Mini Program DAU exceeding 6 million, resulting in a cross-channel de-duplicated DAU of about 13-14 million; the standalone App MAU is approximately 28 million, with total cross-channel MAU nearing 40 million. According to expert calculations, by the end of FY2026, Afu aims for MAU to exceed 50 million, DAU around 20 million, and connected health monitoring device users around 8 million, supported partly by continued marketing investments. The current user structure is concentrated among urban white-collar workers and middle-class individuals aged 25-45, with strong penetration in higher-tier cities. Subsequently, it hopes to sink into lower-tier cities and county markets, where high-quality medical resources are relatively scarce, and the incremental value of AI pre-consultation and health management services is greater. In terms of retention strategy, Afu aims to increase its 30-day retention rate to above 40%. The key measure is to extend from low-frequency, occasional symptom inquiries to continuous health management. Connected user health monitoring devices are seen as an important lever to enhance stickiness—routine monitoring of indicators such as weight, body fat percentage, and blood pressure, synchronized with Afu, creates repeated usage scenarios, embedding Afu into users' daily health habits. On commercialization, experts believe Afu is still in its early stages, with management prioritizing user scale and health engagement over aggressive monetization. The current revenue model mainly involves earning transaction commissions while connecting third-party physical examination service providers, pharmacies, and insurance companies. Commission levels include: physical exam packages approximately 20-25%, OTC drugs 8-10%, prescription drugs approximately 5%, and insurance products 5-10%. In the medium to long term, Afu can expand monetization through membership subscriptions, customized networked health hardware, and B2B AI services for pharmaceutical companies and medical institutions. However, it remains cautious about advertising, especially paid rankings or sponsored recommendations, believing this would undermine the neutrality and credibility of AI health advice. In the competitive landscape, experts do not believe Afu will substantially replace JD Health (6618.HK, Buy) or AliHealth (0241.HK, Not Rated) in the short term. Afu mainly follows an asset-light route; Ant is unlikely to build completely self-operated drug retail or medical service infrastructure, making Afu more like an AI-driven healthcare entry point: AI identifies and clarifies user needs, vertical health records enhance personalization, and third-party service providers complete transaction fulfillment. Therefore, experts judge that Afu is more likely to cooperate with existing internet healthcare platforms and service providers rather than compete directly.

Analysis framework

Nomura's analysis starts with an expert call, unfolding a progressive breakdown around a single emerging product (Afu): 'Strategic Positioning — Technical Barriers — User Metrics — Retention Mechanism — Monetization Path — Competitive Relationship.' Step one clarifies that Afu is not a simple Q&A tool but a long-term strategy combining three modules aimed at full-cycle family health management. Step two demonstrates its differentiation from general AI assistants through technical architecture and data sources. Step three quantifies user expansion progress using specific DAU, MAU, retention rate targets, and marketing investments. Step four breaks down revenue structures such as transaction commissions and deduces future monetization directions. Step five concludes with judgments on competitive relationships with already covered targets like JD Health and AliHealth, finally returning to Nomura's ratings and valuation frameworks for these related listed companies. The entire analysis is primarily qualitative, with data coming from expert estimates, without providing independent profit forecasts or valuation models for Afu.

Methodology notes

• Industry/Industrial Analysis FrameworkSupply and Demand Framework

  Explains the incremental value of Afu sinking into lower-tier cities and county markets from the perspective of insufficient medical resource supply and unmet demand.

  High-quality medical resources in lower-tier areas are relatively scarce, while residents' health consultation and management needs objectively exist. The supply of AI pre-consultation and health management services happens to fill this gap, so the sinking market is viewed as a natural direction for Afu's user expansion.

• Company Fundamentals and Financial FrameworkProfit Quality Analysis

  Judges the stage and quality of Afu's current commercialization through transaction commission rates and revenue structure breakdown.

  The report lists corresponding commission rates for Afu's revenue by categories such as physical exams, OTC drugs, prescription drugs, and insurance, indicating that its current monetization mainly relies on lightweight matchmaking transactions rather than heavy-asset self-operation, and commercialization is still in its early stages.

• Competition and Strategy FrameworkMoat / competitive advantage

  Afu's differentiation comes from the combination of general models + specialized medical models + professional medical data + proactive questioning ability + medical risk control layers.

  Compared to general AI assistants, Afu possesses more focused medical data and stronger risk control, constituting a differentiated advantage relative to platforms like JD Health in the 'front-end user attraction' phase; however, its asset-light positioning makes it difficult to form direct confrontation with heavy-asset platforms in service fulfillment.

• Industry/Industrial Analysis FrameworkUpstream, Midstream, and Downstream Transmission in the Industry Chain

  As an 'AI healthcare entry point', Afu is only responsible for front-end demand identification and traffic diversion, while backend transactions are completed by third-party service providers.

  This division of labor reflects the positioning differences in different links of the internet healthcare industry chain: AI is responsible for information matching and user stickiness, while hospitals, pharmacies, physical examination institutions, and insurance companies are responsible for fulfillment. Afu's asset-light model determines that it leans more towards being an upstream traffic entry point rather than a downstream service provider.

• Cycle and Prosperity FrameworkPenetration Rate S-Curve

  Characterizes the development stage Afu is in using user metrics such as MAU, DAU, and number of connected device users.

  The report uses data such as de-duplicated DAU of approximately 13-14 million, MAU nearing 40 million, and FY2026 targets to position Afu as still in the early climbing stage of user penetration. The core contradiction in this stage is scale expansion and retention improvement, rather than profitability.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• JD Health (6618.HK)

  Afu will not substantially replace JD Health in the short term; the two are more likely to be cooperative rather than directly competing; JD Health, as one of the existing internet healthcare platforms, is directly named for comparison.

  Strengths

  Heavy-asset self-operated drug retail and medical service capabilities, possessing complete fulfillment infrastructure; Nomura gives a Buy rating with a target price of HKD 67.

  Weaknesses

  Faces diversion risks from new entrants like Afu in dimensions such as AI front-end customer acquisition and personalized health management.

  Comparison

  Compared to Afu's asset-light entry model, JD Health is better positioned to handle actual transaction fulfillment; Afu lacks backend services, while JD Health lacks an AI-driven user interaction entry point.

  Risks

  Downside risks mentioned in the report include: regulatory risks in the online drug/medical services industry, intensified competition from emerging players, and slower-than-expected monetization of medical services.

• Alibaba (BABA.US)

  Afu is launched by Ant Group, an affiliate of Alibaba; Alibaba supports Afu at the model layer (Tongyi Qianwen) and ecosystem layer; Alibaba itself is rated Buy with a target price of USD 178.

  Strengths

  Possesses Alibaba Cloud and the Tongyi Qianwen model ecosystem, capable of providing a general model foundation for Afu.

  Weaknesses

  The report does not elaborate on disadvantages of Alibaba relative to Afu.

  Comparison

  As an affiliate of Ant, Alibaba has a different positioning in the healthcare track compared to JD Health; the former leans towards technology and ecosystem empowerment, while the latter leans towards service fulfillment.

  Risks

  Downside risks for Alibaba mentioned in the report include: profit pressure brought by increased investment, and regulatory risks related to payment and internet finance potentially impacting its value in Ant.

• AliHealth (0241.HK)

  Used to compare the competitive impact on Afu as an existing internet healthcare platform; Nomura gives it a Not Rated status.

  Comparison

  Like JD Health, it belongs to internet healthcare platforms potentially affected by Afu's competition, but the report judges that Afu will not substantially replace these platforms in the short term.

Key data

• Afu Standalone App DAU8-9 millionExpert estimate, excluding Mini Programs
• De-duplicated Cross-channel DAUApproximately 13-14 millionIncludes multiple channels such as App and Mini Programs
• Total Cross-channel MAUNearly 40 millionStandalone App MAU approx. 28 million
• FY2026 End MAU TargetExceed 50 millionDAU target approx. 20 million, connected health device users approx. 8 million
• 30-Day Retention Rate TargetIncrease to above 40%Improving retention through continuous health management scenarios
• Physical Exam Package Commission RateApproximately 20-25%One of the current main monetization sources
• OTC Drug Commission Rate8-10%Prescription drug commission approx. 5%
• Insurance Product Commission Rate5-10%Earning matchmaking income by connecting insurance companies
• JD Health Rating and Target PriceBuy, HKD 67Based on 21x FY2026 adjusted P/E valuation; Current price HKD 38.94 (2026-08-07)
• Alibaba Rating and Target PriceBuy, USD 178Derived from sum-of-the-parts valuation; Current price USD 128.41 (2026-08-07)

Impact & implications

For related listed companies, the research report believes that Afu will not have a substantial impact on JD Health and AliHealth in the short term, and is more likely to cooperate with these platforms as an AI-driven front-end entry point, which is relatively friendly to the stability of the existing internet healthcare landscape. For the Ant system, Afu is an asset-light lever for its healthcare layout. Leveraging Alipay's existing healthcare infrastructure and Alibaba's model ecosystem, it has the opportunity to occupy user mindshare in the popularization of medical AI applications. For the industry, AI medical assistants are evolving from single Q&A to continuous health management, and the core of user stickiness is shifting to 'whether it can enter daily life scenarios,' which may drive the competition in medical AI from traffic acquisition to retention and scenario depth.

Risks

• Regulatory risks in the online drug and medical services industry
• Intensified competition due to the entry of emerging players
• Slower-than-expected monetization speed of healthcare services

What to watch

• Whether Afu's actual growth in DAU, MAU, and connected device users reaches the FY2026 end targets
• Whether the 30-day retention rate can be increased to above 40%
• Whether new monetization paths beyond transaction commissions (membership subscriptions, networked hardware, B2B AI services) are implemented
• How the cooperative relationship between Afu and platforms such as JD Health and AliHealth evolves