OpenAI, Anthropic US House Committee Staff on New AI Models
The sessions represent some of the earliest direct engagements between lawmakers and leading AI developers specifically focused on the threat landscape emerging from cutting-edge models.
The two firms appeared before congressional staff in separate closed-door sessions last Thursday. Anthropic has withheld public access to its Mythos Preview model over concerns about its potential to exploit security vulnerabilities, while OpenAI has opted for a tiered-release framework for its GPT-5.4-Cyber model, calibrating access based on risk assessment.
A committee aide described the meetings as "proactive engagement with these companies on recent frontier model developments," particularly their impact on critical infrastructure cybersecurity.
The aide noted that discussions also touched on a recent White House memo alleging that China is conducting "industrial-scale" efforts to replicate and distill American AI systems — a development that has sharpened urgency on Capitol Hill.
House Homeland Security Chair Andrew Garbarino, who has convened a series of private roundtables with technology and AI executives, has also steered the committee through multiple hearings examining generative AI's national security implications, including the growing threat of state-sponsored cyberattacks.
"Productive partnerships between industry and government are essential to help us stay ahead of the evolving threat landscape, ensure the government is prepared to securely harness AI for its defensive capabilities, and support and protect American AI development as adversaries like China seek to gain an advantage by any means," Garbarino told media.
A separate briefing held the same week delivered an even more visceral warning. Lawmakers were shown live demonstrations of jailbroken AI — models deliberately stripped of their safety constraints — and walked through scenarios in which such tools could be weaponized to facilitate mass casualty events, including school shootings and bombings.
The demonstrations left a visible impression on those present.
"What I just saw in there, with just a short amount of time typing in questions, is very scary. These models are very powerful," said Rep. August Pfluger following the session.
"We see how powerful it is, and it should be used for good, but guardrails need to be attached... Congress and the executive branch need to work with our industry partners to make sure that we keep kids safe," Pfluger added, calling for coordinated action between government and the private sector.
Legal Disclaimer:
EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
