The Food and Drug Administration launched Elsa, an artificial intelligence assistant, on Monday to mixed reviews from employees who describe the tool as premature and error-prone.

FDA Commissioner Marty Makary touted the swift deployment, stating, “I set an aggressive timeline to scale AI agency-wide by June 30. Today’s rollout of Elsa is ahead of schedule and under budget, thanks to the collaboration of our in-house experts across the centers.”

The generative AI system aims to streamline clinical protocol reviews, scientific evaluations and inspection targeting. FDA Chief AI Officer Jeremy Walsh proclaimed, “Today marks the dawn of the AI era at the FDA. (W)ith the release of Elsa, AI is no longer a distant promise but a dynamic force enhancing and optimizing the performance and potential of every employee.”

Internal Skepticism Surfaces

Testing showed major flaws when staff queried Elsa about FDA-approved products Monday, receiving incorrect or partially wrong responses, NBC News reported.

“Makary and DOGE think AI can replace staff and cut review times, but it decidedly cannot,” one employee told Stat news. Another staffer expressed concern about missing safeguards: “I’m not sure in their rush to get it out that anyone is thinking through policy and use.”

Built on Anthropic’s Claude technology by Deloitte, the system cost $13.8 million since 2020, with an additional $14.7 million contract awarded in April. The tool evolved from individual center pilots, with CDER-GPT selected for agency-wide expansion and rebranding as Elsa.

Read more at Ars Technica