There are more AI health tools than ever—but how well do they work?
AI Summary
In early 2026, Microsoft launched Copilot Health within its Copilot app, allowing users to connect medical records and ask health questions, joining Amazon's newly expanded Health AI (previously restricted to One Medical members), OpenAI's ChatGPT Health (released January 2026), and Anthropic's Claude, which can access health records with permission. Microsoft reports receiving 50 million health questions per day on Copilot, with health being the most popular topic on its mobile app, underscoring the enormous consumer demand driving these launches. According to Technology Review, all six academic experts interviewed expressed concern that these tools are being released without independent third-party safety evaluations, despite the high-stakes nature of health applications including triage, diagnosis, and treatment guidance. A recent Mount Sinai study found ChatGPT Health sometimes over-recommends care for mild conditions and fails to identify emergencies, while a separate Oxford Internet Institute study found that non-expert users assisted by LLMs correctly identified medical conditions only about one-third of the time when tested with realistic scenarios. OpenAI has developed its own HealthBench evaluation framework and currently holds the top score on Stanford's MedHELM benchmark, though Stanford professor Nigam Shah acknowledged MedHELM only assesses single-turn responses rather than multi-turn medical conversations. Google's AMIE chatbot demonstrated diagnostic accuracy comparable to physicians in a recently published human study, but Google stated it does not plan a public release until equity, fairness, and safety limitations are addressed; Google did separately announce a Gemini-powered AI assistant for its Health100 platform, being built in partnership with CVS.
Why it matters
The rapid simultaneous deployment of AI health tools by Microsoft, Amazon, OpenAI, and Google signals a major competitive land-grab in the consumer health AI sector, a market driven by documented demand of tens of millions of daily queries and structural gaps in healthcare access. The absence of standardized, independent third-party benchmarks represents a regulatory and reputational risk for these companies, as scrutiny from researchers and institutions like Mount Sinai and Stanford could accelerate calls for formal oversight, potentially affecting the pace and terms of future product rollouts. The Google-CVS Health100 partnership and OpenAI-Microsoft collaboration illustrate how AI companies are deepening ties with established healthcare and retail players, pointing to a broader trend of cross-sector integration that may reshape competitive dynamics across both the technology and healthcare industries.
Scoring rationale
The article covers major AI companies (Microsoft, Amazon, OpenAI, Google, Anthropic) launching consumer health AI products with significant market implications, but focuses primarily on safety and evaluation debates rather than direct financial or market-moving developments.
Impacted tickers
This summary was generated by AI from the original article published by MIT Technology Review AI. AIMarketWire does not provide trading advice. Always refer to the original source for complete reporting.