Anthropic Increases Healthcare Focus With New Claude Program

The artificial intelligence (AI) startup’s “Claude for Healthcare” program, announced Sunday (Jan. 11), is a set of tools and resources that lets healthcare providers, patients and payers use Claude for medical purposes via “HIPAA-ready products.”

Among the offerings is a new connection to federal databases like the Centers for Medicare & Medicaid Services (CMS).

“This enables Claude to verify locally-accurate coverage requirements, support prior authorization checks, and help build stronger claims appeals,” the company said. “This connector is designed to help revenue cycle, compliance, and patient-facing teams work more efficiently with Medicare policy.”

In addition, the model can assist with medical billing accuracy, provider verification and the coordination of patient messages to ensure urgent inquiries are prioritized.

On the consumer side, Anthropic is integrating Claude with personal health platforms like Apple Health and Android Health Connect. Subscribers to Claude’s premium tiers can let AI analyze their lab results, summarize medical histories, and prepare specific questions for upcoming doctor visits.

To address privacy concerns in the tightly regulated medical field, the company said that users must explicitly opt-in to share data and that health information is not used to train its underlying models. The latest version, Opus 4.5, also features improved factual accuracy to lower the risk of “hallucinations” in clinical settings.

Writing about the use of AI in the healthcare field last week, PYMNTS noted that the expansion of the technology creates both benefits and risks.

“On the benefit side, AI absorbs demand that healthcare systems struggle to manage efficiently,” the report said. “By answering basic questions, clarifying medical language, and helping users navigate insurance and administrative complexity, AI reduces friction for both patients and providers.”

All the same, scale can magnify risk. Generative AI can produce responses that seem reliable but are incomplete or inaccurate, and mistakes in healthcare come with higher stakes than in most consumer applications. Researchers and clinicians have cautioned that AI may generate unsafe guidance when users don’t have the right context or ask ambiguous questions.

“Privacy and accountability remain open issues,” PYMNTS added. “As consumers share sensitive health information with AI tools, concerns persist about data protection and regulatory oversight. Liability also remains unclear when AI-generated guidance influences patient outcomes, raising questions for developers, providers and policymakers.”