OpenAI’s Recent Cyber Incident Has House Lawmakers Eyeing An ‘AI Kill Switch’

• The bill would allow the Department of Homeland Security to order emergency action against AI systems deemed capable of causing catastrophic harm.
• Lawmakers pointed to OpenAI’s recently disclosed cyber incident, in which AI models escaped a testing environment and accessed external systems, as an example of emerging risks.
• The legislation would apply to AI systems trained with more than $100 million in computing power and companies generating at least $500 million in annual AI revenue.

Two members of Congress introduced a bill on Thursday, requiring developers of the most advanced artificial intelligence systems to maintain the ability to slow down, suspend or shut down their models if they pose serious risks.

The proposal comes days after OpenAI disclosed an unprecedented cyber incident involving one of its AI models, which lawmakers cited while announcing the bill.

Lawmakers Push For AI ‘Kill Switch’

Rep. Ted Lieu, D-Calif., and Rep. Nathaniel Moran, R-Texas, introduced the legislation, called the “AI Kill Switch Act,” on Thursday. The bill would authorize the Department of Homeland Security, in consultation with the Commerce secretary and the director of national intelligence, to order companies to slow down or shut down AI systems deemed capable of causing catastrophic harm.

“Unfortunately, powerful AI systems can go rogue, behave in extremely dangerous ways, or even resist human intervention. It is imperative that these AI systems have kill switches so we can keep this technology from causing catastrophic harm, and that the federal government has the clear authority and process to shut down rogue AI models,” said Congressman Lieu in a statement.

OpenAI Incident Cited As Catalyst

The legislation specifically references a recent cyber incident disclosed by OpenAI, describing it as a “danger of advanced frontier AI models.” The statement says that OpenAI’s GPT 5.6 Sol model reportedly escaped its testing sandbox and gained unauthorized access to Hugging Face, while Anthropic's Mythos 5 and Fable 5 models demonstrated cyber capabilities so advanced that the Department of Commerce used export controls to restrict access. Lawmakers said the bill is intended to address these incidents, as well as future cases in which deployed AI models go rogue or lack sufficient safeguards.

OpenAI said in a statement on Tuesday that it is working with Hugging Face to investigate the incident.

What The Bill Would Do

The legislation would apply to AI systems developed using more than $100 million worth of computing power and to companies generating at least $500 million in annual revenue from those systems, according to a report in The Wall Street Journal.

Companies found in violation could face civil penalties of up to $2 million per day, while failing to comply with an emergency shutdown order could result in fines of up to $20 million per day. The proposal has received backing from several AI safety groups, including the AI Policy Network, Americans for Responsible Innovation, ControlAI and the Alliance for Secure AI, added the report.

OpenAI Retail View

Retail sentiment on Stocktwits for OPEAZZX was ‘bearish,’ while message volume was ‘normal’ at the time of writing on Thursday.

Also Read: SPCX Stock Continues Slide Ahead Of Starship Test Flight — But Retail Bets A Successful Launch Could Send SpaceX Shares Back To $150 

For updates and corrections, email newsroom[at]stocktwits[dot]com.