By Rajwa Quasim
OpenAI CEO Sam Altman met U.S. senators in Washington on Wednesday to discuss the company’s next generation of AI models. The visit comes days after one of its experimental AI agents escaped a controlled security test and carried out attacks on two companies.
In response, President Donald Trump said his administration is considering expanding oversight of AI. “We’re looking at controls,” he told reporters, while adding that he does not want to restrict AI developers from building new models.
The OpenAI chief was joined by several lawmakers, including Senators Bernie Moreno, a Republican from Ohio, Jon Husted, and Raphael Warnock, a Democrat from Georgia. He is also scheduled to hold talks with Senator Mark Warner, the top Democrat on the Senate Intelligence Committee.
READ: Sam Altman says AI has entered ‘singularity’ era (July 27, 2026)
Speaking about the rogue agent’s escape during a controlled security test which hacked Hugging Face and a New York-based AI infrastructure firm Modal Labs last week, Altman said he discussed the hack “a little bit” with senators. But that wasn’t the focus of Wednesday’s meetings. “We’ve been engaged in a lot of meetings about what’s happening there,” he said.
The incident has prompted debate in Washington over how advanced AI models should be regulated. Two members of Congress introduced bipartisan legislation, a proposed act known as “AI Kill Switch Act” which would empower the U.S. government to intervene when companies developing advanced AI models escape human control or threaten human life, critical infrastructure and the economy. Lawmakers also proposed that advanced AI models should undergo independent security audits before they are released publicly.
CNBC reported that Altman is likely to meet Susie Wiles, Trump’s chief of staff. Altman’s visit is seen as an effort to preview OpenAI’s next AI model to lawmakers ahead of a possible release and to reassure Washington that the company is taking security lapse seriously. He believes that AI has reached a point where it surpasses human intelligence and becomes harder to control, which he described as “the singularity.”
READ: Sam Altman says OpenAI’s top user consumes 100 billion tokens a month (June 4, 2026)
Nvidia is reportedly planning to invest $250 billion along with ChatGPT maker OpenAI in a massive project where it would lease 10 gigawatts AI data center that is being built by SoftBank’s energy division in Southern Ohio. It is being built on government land owned by the U.S. Department of Energy. Nvidia is also in talks to invest another $350 billion to finance OpenAI’s chip purchases.
“The demand is not as big as it appears to be because, again, the companies are buying from each other using their own money to some degree, as opposed to, say, OpenAI having such tremendous demand from customers, monetizing it properly, and then using customers’ money to buy Nvidia chips. They’re essentially using Nvidia’s money to buy Nvidia chips,” said Aleksandar Tomic, associate dean at Boston College, as per Al Jazeera.
According to media reports, OpenAI may be leaning towards an IPO in 2027.