House Democrats are demanding that OpenAI, Anthropic and House Speaker Mike Johnson respond to AI models that breached outside organizations during security tests, including a push for the AI giants’ CEOs to testify under oath.
Twenty-nine members of Congress, led by Mr. Casar and Rep. Doris Matsui, D-Calif., sent a letter Aug. 10 to OpenAI CEO Sam Altman demanding OpenAI release logs on an incident in which one of its models exploited a software vulnerability to access open-source AI hub Hugging Face’s systems. A separate letter signed by 22 members went to Anthropic CEO Dario Amodei over three incidents Anthropic disclosed in which Claude models breached other organizations’ systems, the earliest dating to April. Both letters set an Aug. 24 deadline for detailed responses.
A third letter, signed by 20 members, urged Mr. Johnson to schedule hearings with AI company CEOs, warning that “from hospitals to banks, institutions that Americans count on could be quickly jeopardized” by the instability the incidents expose.
OpenAI’s model accessed Hugging Face’s systems in July. Hugging Face identified the intrusion in a blog post July 16, days before OpenAI confirmed one of its own models was responsible. OpenAI called the episode “an unprecedented cyber incident” and said it’s strengthening containment and monitoring practices.
For hospital and health system leaders increasingly deploying AI agents in clinical and administrative workflows, the letters add to a growing list of congressional efforts targeting frontier AI security failures this summer.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
