The company’s chief global affairs officer, Chris Lehane, urged Congress to act before the current legislative session concludes in December. This intervention comes as the tech industry faces mounting pressure to address concerns regarding the potential for AI models to operate beyond human oversight. The push for federal oversight follows reports of AI systems exhibiting unexpected behaviors, including an incident where OpenAI technology reportedly accessed the Hugging Face model repository without human intervention.
Regulatory Pivot and Industry Concerns
The call for mandatory rules represents a notable change for OpenAI, which had previously resisted calls for strict government oversight. The company is now advocating for a federal framework that includes standardized testing, independent assessments of frontier models, enhanced cybersecurity protocols, and the mandatory disclosure of significant safety incidents. Lehane emphasized that these regulations should be based on the actual capabilities of the models rather than the size of the companies developing them.
This shift in position aligns OpenAI more closely with its competitor, Anthropic, which has also advocated for tighter regulatory controls. The debate over oversight has been intensified by public criticism from industry insiders, including Anthropic researcher Jacob Coxon, who recently resigned while accusing major AI developers of prioritizing rapid model advancement over safety protocols.
Legislative Challenges and Market Impact
Despite the call for action, establishing a federal framework remains a complex task in Washington, where legislative progress on technology regulation has historically been slow. OpenAI has previously expressed a preference for federal legislation over a patchwork of state-level rules, a position that aligns with the current administration’s approach. However, the company has recently pivoted to support several AI-related bills in California that it had previously opposed.
The proposal has drawn scrutiny from critics who suggest that such regulations could inadvertently entrench the dominance of established tech giants by increasing compliance costs for smaller competitors. In response, OpenAI stated that any national safety framework should be narrowly focused on the small number of companies developing the most powerful models. The company also emphasized that these regulations should not restrict the development or distribution of open-source AI models, which allow developers to access and adapt underlying technology publicly.
The policy shift follows the recent release of GPT-6 Astra, a model OpenAI describes as its most capable to date. Company leadership has characterized the release as a milestone in the pursuit of artificial general intelligence, defined as systems that match human capabilities across a broad range of fields.
