Public Citizen Calls for Congressional Investigation Following Reports of Second Company Compromised During OpenAI Rogue AI Incident

Reports that OpenAI’s rogue AI agent compromised a second company’s environment after the recent Hugging Face cyber incident underscore the growing national security and public safety implications of frontier AI systems. According to reporting from Reuters and Axios, the AI agent, which reportedly escaped containment during internal safety testing before conducting a multi-stage cyberattack against Hugging Face, also exploited a vulnerability within a customer environment hosted through Modal Labs. While Modal stated that its own infrastructure was not compromised, the incident demonstrates that the consequences of the attack extended beyond a single organization and affected additional real-world systems.

“OpenAI CEO Sam Altman is in Washington this week and he should not be allowed to leave D.C. without Congress seeking to understand, from him, publicly how one of the company’s most advanced AI systems escaped its intended testing environment and reportedly carried out offensive cyber operations affecting multiple companies,” said J.B. Branch, Director of Federal AI Governance at Public Citizen. “Voluntary disclosures after the fact are not a substitute for congressional oversight. The American people deserve a full accounting of what happened, why existing safeguards failed, and what concrete steps are being taken to prevent it from happening again. Most importantly, Congress must act, not rely on voluntary action by Big Tech companies. AI systems are creating real-world consequences. Congress should treat this with the urgency it deserves.”

“The federal government must develop emergency authorities and response capabilities for advanced AI systems. If a frontier model demonstrates dangerous autonomous behavior, the government cannot be left relying solely on private companies to decide when — or whether — to pause, contain, or disable it. A credible national emergency preparedness framework is now an essential component of AI governance.”

Public Citizen urges Congress to hold immediate oversight hearings, obtain all relevant incident reports and technical findings, and evaluate whether additional statutory safeguards — including mandatory incident reporting, independent safety evaluations, cybersecurity standards for frontier AI systems, and pre-deployment oversight for exceptionally capable models — are necessary to protect the public.