HHS tells employees to abandon Claude AI, as Trump feuds with developer Anthropic
As the federal government’s dispute with AI company Anthropic escalates, largely stemming from safeguards the company demanded for how its technology can be deployed, the U.S. Department of Health and Human Services (HHS) has now banned the use of its popular generative large language model, Claude.
The dispute stems from Anthropic—which has contracts with multiple federal agencies, including the Department of War—demanding an assurance that its artificial intelligence will never be deployed in autonomous weapons or as a means to perform mass surveillance of American citizens.
President Donald Trump’s administration has since been looking to blacklist the company from official government use, which led to HHS implementing a rule that instructs federal healthcare workers to seek alternatives, replacing Claude with OpenAI’s ChatGPT or Google’s Gemini.
“As of today, users will no longer be able to log in to or access Claude through the HHS enterprise environment,” read an email sent to HHS employees. “Please discontinue any ongoing use of the platform and transition your work to one of the Department’s other approved enterprise AI solutions.”
This comes after a full rollout of Claude was announced in December, impacting all agencies under HHS—mainly the U.S. Food and Drug Administration (FDA).
The agency had set out to use Claude for speeding up its drug review process, although it isn’t clear if regulators were currently using the AI for that purpose, given that the push for adoption was ongoing.
Notably, the FDA used Claude as the basis for its own platform, called Elsa, which was meant to accelerate new drug and expanded use approvals.
It’s unclear if the switch to ChatGPT and Gemini will delay any plans HHS departments had of deploying AI, something the Trump administration has pushed as a priority.
‘No substitute,’ for Claude FDA employee says
Speaking to NOTUS—the outlet that first broke the story of HHS abandoning Anthropic's AI and revealed the above internal email—an anonymous FDA employee said the decision to make the switch to another LLM platform “will basically wipe out 18 months of efforts.”
“For serious scientific work, study reviews and regulatory writing or R-coding, there is no substitute,” NOTUS quoted the employee as saying. “For an entire year we have been training reviewers on writing prompts and using the [LLM], and finally it looks like the AI was catching on.”
It’s unclear if HHS will be permanently making the shift, or if this is a negotiating tactic by the Trump administration over the disputes related to homeland security.
For more, read the full coverage from NOTUS by clicking here.
Chad is an award-winning writer and editor with over 15 years of experience working in media. He has a decade-long professional background in healthcare, working as a writer and in public relations.