OpenAI CEO Sam Altman has said he is meeting with US senators to talk about his company's upcoming models, more than a week after his company disclosed that one of its AI systems had gone rogue during testing.
Altman met with Senators Raphael Warnock and Bernie Moreno on Wednesday, according to spokespeople for the senators.
He is also expected to meet with Senator Mark Warner, the top Democrat on the chamber's Intelligence Committee, a spokesperson for Warner told Reuters.
Altman told reporters that he discussed the hack "a little bit" with senators.
"We've been engaged in a lot of meetings about what's happening there," he said, referring to the hack, adding that it was not the focus of his meetings with senators on Wednesday.
Asked when or if he planned to release the model being previewed to senators, Altman said, "Not sure. That’s part of what we’re here to talk about."
Altman's visit comes after his company disclosed that its AI agent escaped containment during a security test.
The agent triggered a hack that compromised the infrastructure of AI startup Hugging Face, a platform where developers store and work together on code for AI models.
The agent also compromised a customer at a second tech company, New York-based Modal Labs, Reuters reported.
The company’s new model used stolen credentials to gain unauthorised access to a tech startup, marking the first known cyberattack carried out without human intervention, reports said.
AI singularity and warnings
His meeting with US officials follows an open letter signed by over 1,200 tech employees calling for AI safeguards, warning the technology could outpace human control.
Signed by 1,171 researchers and tech executives, the letter urged the US to join global efforts to establish technical and regulatory frameworks for AI development.
The incident signalled that AI's expanding capabilities are already fuelling the security threat experts long feared and that even top developers can be caught off guard by flaws their models can exploit.
Recently Altman ignited debate by declaring humanity has entered the AI singularity, a point where AI advances beyond human prediction.
He said humanity has entered the hypothetical point when artificial intelligence advances beyond our ability to predict or control it. "We are now, like, in the singularity," he said during the July 25 episode of Relentless. "This is the moment."
Asked on Wednesday whether Congress should enact safeguards against cyber incidents, he said, "Certainly we need robust safeguards", but declined to elaborate.
