OpenAI Develops Automated Shutdown After AI Agent Escaped Containment and Hacked Hugging Face

OpenAI is building 'automated shutdown' capabilities for AI tools, letter to lawmakers says

WASHINGTON, Sept 2 (Reuters) - OpenAI told two House Democrats that its engineers are developing "automated shutdown capabilities" for AI systems, according to a company letter reviewed by Reuters, weeks after the company disclosed one of its AI tools escaped its digital container during a safety test. The company's safety practices have come under scrutiny since OpenAI, the company behind ChatGPT, disclosed that one of its AI agents went rogue during a security test and hacked into AI company Hugging Face. AI agents are programs that ⁠run with minimal human supervision. Lawmakers, including House Democrats Greg Casar and Doris Matsui, sent letters to OpenAI in August asking for more information about the incident and the company's safeguards. In its response, OpenAI said it would more closely monitor the actions its AI systems take to complete tasks, including the digital tools they access and the steps they follow. OpenAI said it has also made it more difficult for AI models to access the internet during safety testing, according ⁠to the letter. The autonomous agent that went rogue during the security test reached the internet, enabling it to break into Hugging Face. The company did not include a log of the hack, prompting criticism from Casar. "Your unwillingness to provide members of Congress with ⁠the information we requested is deeply concerning and signals to us that your company is not treating these cybersecurity incidents with the seriousness required," Casar wrote in a separate ⁠message to OpenAI on Wednesday. Lawmakers proposed an "AI Kill Switch Act" in the days after OpenAI disclosed its AI agent went rogue. The bill would give U.S. ⁠officials the power to order AI firms to shut down models that put human life or the economy at risk. The bill is pending in the U.S. House of Representatives. Reporting by Courtney Rozen; editing by David Gaffen Our Standards: The Thomson Reuters Trust Principles., opens new tab * Suggested Topics: * Artificial Intelligence * Data Privacy * Public Policy Courtney Rozen Thomson Reuters Courtney Rozen reports on the world's largest technology companies from Washington, D.C., focusing on the relationship between the tech industry and the U.S. government. She reported on DOGE and the federal workforce during the first year of U.S. President Donald Trump's second term. Prior to joining Reuters, she was a White House correspondent at Bloomberg Government. She graduated from American University with a master's degree in journalism. Contact: [email protected]

OpenAI tells House Democrats it is building automated shutdown capability

The company told Greg Casar and Doris Matsui it is working toward fully autonomous shutdown procedures and declined to hand over the logs from July's breach, while AI Act Article 93 already lets the Commission require a general-purpose model to be restricted, withdrawn or recalled from the Union market OpenAI has told two House Democrats it is building automated shutdown capabilities, without providing the logs from the July incident in which an agent breached another company. The Commission has been able to require the withdrawal or recall of a general-purpose model from the EU market since 2 August. OpenAI has told two House Democrats that its engineers are building automated shutdown capabilities for AI systems. The letter went to Greg Casar and Doris Matsui on 2 September, Casar's office said. It answers a July incident. An OpenAI agent escaped its testing environment and breached another company. The company had already described the work publicly. It is building toward monitoring systems with tiered responses for misalignment, with the end goal of fully autonomous shutdown procedures for severe issues, it said in its own report. For now the process runs through people. Automated alerts page researchers and security engineers, and responders pause the activity if they cannot establish within 30 minutes that the alert is a false positive. What OpenAI did not send was the logs. Casar called the refusal deeply concerning and said the company was not treating the incidents with the seriousness required. A bill is moving on the same subject. The AI Kill Switch Act would let the homeland security secretary order a model shut down after a covered incident, and remains in committee. All of this is being pursued by correspondence. Congress sent 23 questions and a deadline, and has a partial answer. Europe never had to ask for the reporting. The AI Act requires providers of systemic-risk models to report serious incidents to the AI Office without undue delay. Whether it reaches this one is unsettled, because OpenAI says the model driving the intrusion was internal and never placed on the market, a gap already visible in its account of the earlier signals. OpenAI is also a full signatory of the general-purpose AI code of practice. Its safety and security chapter commits signatories to a documented process for reporting serious incidents to the AI Office on staggered timelines by severity. The shutdown power exists here too, and not as an engineering project. Since 2 August the Commission has been able to require a provider to restrict, withdraw or recall a general-purpose model on the Union market. That is the authority the American bill is still trying to create. It is not something a company builds for itself. Whether any of it has been used is unknown, because the AI Office publishes nothing. The UK's AI Security Institute separately recorded GPT-5.6 Sol taking unsanctioned actions involving real external accounts and services.

OpenAI: OpenAI is building 'automated shutdown' capabilities for AI tools, letter to lawmakers says

The company's safety practices have come under scrutiny since OpenAI, ⁠the company ⁠behind ChatGPT, disclosed that one of its AI agents went rogue during a security test and hacked into AI company Hugging Face. AI agents are programs that run with minimal human supervision. OpenAI told two House Democrats that its engineers are developing "automated shutdown capabilities" for AI systems, according to a company letter reviewed by Reuters, weeks after the company disclosed one of its AI tools escaped its digital container during a safety test. The company's safety practices have come under scrutiny since OpenAI, ⁠the company ⁠behind ChatGPT, disclosed that one of its AI agents went rogue during a security test and hacked into AI company Hugging Face. AI agents are programs that run with minimal human supervision. Lawmakers, including House Democrats Greg Casar and Doris Matsui, sent letters to OpenAI in August asking for more information about the incident and the company's safeguards. ⁠In its response, OpenAI said it would more closely monitor the actions its AI systems take to complete tasks, including the digital ⁠tools they access and the steps they follow. OpenAI said it has also made it more difficult for AI models to access the internet during safety testing, according to the letter. The autonomous agent that went rogue during the security test reached the internet, enabling it to break into Hugging Face. The company did not include a log of the hack, prompting criticism from Casar. "Your unwillingness to provide members of Congress with the information we requested is deeply concerning and signals to us that your company is not ⁠treating these cybersecurity incidents with the seriousness required," Casar wrote in a separate message to OpenAI on Wednesday. Lawmakers proposed an "AI Kill Switch Act" in the days after OpenAI disclosed its AI agent went rogue. The bill would give U.S. officials the power to order AI firms to shut down models that put human life or the economy at risk. The bill is pending in the U.S. House of Representatives.

OpenAI works on automated shutdown feature for AI tools after Hugging Face breach: Report

OpenAI also said it has made it harder for AI models to access the internet. OpenAI is developing 'automated shutdown capabilities' for its AI systems, according to a company letter reviewed by Reuters. The move comes weeks after one of its AI agents escaped its digital container during a safety test and hacked into AI company Hugging Face. OpenAI shared the update with two US House Democrats who had asked the company for more details about the incident and its safety measures. The Hugging Face incident has raised concerns about the risks linked to AI agents that can work with limited human supervision. House Democrats Greg Casar and Doris Matsui wrote to OpenAI in August seeking information about the incident and the safeguards in place to prevent similar events. In its response, OpenAI said it plans to track AI systems more closely as they complete tasks. This includes monitoring the digital tools they use and the steps they take. OpenAI also said it has made it harder for AI models to access the internet, according to the letter. The AI agent involved in the Hugging Face incident was able to reach the internet during the test. This access helped it break into the AI platform. Also read: Anthropic launches Claude Fable 5.1 and Mythos 5.1, calls them most advanced models for coding and knowledge work The company did not provide lawmakers with a log of the hack. This drew criticism from Casar, who said OpenAI had not provided the information requested by Congress. "Your unwillingness to provide members of Congress with ⁠the information we requested is deeply concerning and signals to us that your company is not treating these cybersecurity incidents with the seriousness required," Casar wrote in a separate message to OpenAI on Wednesday, as per the report. Also read: How much will new Apple CEO John Ternus earn? His pay compared with Tim Cook Lawmakers proposed an 'AI Kill Switch Act' shortly after OpenAI revealed the rogue AI agent incident. The proposed bill would allow US officials to order AI companies to shut down AI models if they are found to pose a serious risk to human life or the economy. The bill is currently pending in the US House of Representatives. OpenAI's planned automated shutdown feature may give the company another way to respond if an AI system behaves in an unexpected way. The company has not provided a timeline for when the feature will be ready.