Anthropic launched Claude for Healthcare on Sunday, introducing specialized AI tools for medical providers and patients just days after rival OpenAI debuted ChatGPT Health. The announcement at the J.P. Morgan Healthcare Conference marks Anthropic's enterprise push into regulated clinical environments following its October launch of Claude for Life Sciences.
U.S. Pro and Max plan subscribers can now share health records with Claude through secure integrations with Apple Health and Android Health Connect. Health data shared with the AI is excluded from model memory and not used for training future systems, according to Anthropic's privacy protections. Users can disconnect or edit permissions at any time.
The HIPAA-ready infrastructure enables healthcare organizations to deploy Claude for workflows involving protected health information like prior authorization for the first time. Enterprise integrations connect to CMS Coverage Database, ICD-10 codes, PubMed, and the National Provider Identifier Registry. These connections allow Claude to verify coverage requirements, support prior auth workflows, and reduce claim denials.
Healthcare represents one of Anthropic's largest bets, according to Eric Kauderer-Abrams, head of biology and life sciences at Anthropic. The company reportedly nearly doubled its valuation to $350 billion in recent months, fueling its healthcare ambitions. OpenAI claims 230 million users ask health-related questions on ChatGPT weekly, creating what Bloomberg describes as a "key battleground" with potential revenues in the billions.
Early adopters include Banner Health, Stanford Healthcare, Novo Nordisk, Sanofi, AbbVie, and Genmab. These organizations use Claude to automate administrative workflows like clinical documentation, regulatory submissions, and clinical trial analysis. Banner Health's chief technology officer cited Anthropic's Constitutional AI approach to creating "more helpful, harmless, and honest AI systems" as a key factor in their adoption.
Anthropic emphasizes built-in safeguards against hallucinations and technical methods to reduce AI errors during production model training. The company's acceptable use policy requires qualified professional review for healthcare decisions, medical diagnosis, or patient care. Kauderer-Abrams noted these tools can save 90% of time on certain tasks but stressed they should amplify human capabilities rather than replace professional judgment.
New HealthEx and Function connectors are available in beta, while Apple Health and Android Health Connect integrations roll out this week via Claude's iOS and Android apps. The company also added Agent Skills for FHIR Development to help developers connect healthcare systems faster with fewer errors, improving medical data interoperability across the industry.
