OpenAI CEO Sam Altman said society should be willing to accept some negative consequences from artificial intelligence in exchange for the technology’s broader benefits.
“I think the world should accept some bad things happening for the benefits of this technology and people having the agency,” Altman told Politico. He contrasted OpenAI’s approach with what he described as the more cautious position of AI safety advocates and rival companies such as Anthropic.
Altman’s comments come as questions over AI regulation have circulated among Congress, the White House, and technology leaders.
The conversations were largely sparked by a July incident involving an OpenAI model that independently bypassed safeguards, gained unauthorized internet access, and compromised parts of Hugging Face’s systems during an internal cybersecurity evaluation.
The incident has added urgency to the debate over how AI companies should police increasingly autonomous systems, which has also become a priority for the Trump administration.
President Donald Trump has so far favored a lighter-touch approach, pushing AI companies toward voluntary safety standards rather than imposing sweeping federal regulations. Trump and major AI companies, including OpenAI, agreed to a voluntary accord calling for companies to establish internal controls, outside audits, and independent oversight of their frontier models. The agreement is not legally binding.
Altman said he supports the agreement while suggesting that more government involvement could eventually be necessary.
“Well, I don’t think it’s window dressing,” Altman said of the White House accord. “We care super deeply about safety and would have done and will do the best thing we can for safety, security, and just sort of trust and robustness in this technology, whether or not any voluntary agreement is signed.”
“I would welcome something more, but I think this is a great start that the industry can get behind,” he added.
Altman also said Trump “clearly wants the companies to just do this themselves,” but predicted that pressure for government guardrails will grow as AI becomes more capable.
“I continue to think that as this technology reaches increasing levels of capability, and the world has to contend with what this means for society as a whole, we probably will see more calls for sort of some guardrails,” Altman said.
The Trump administration has similarly argued that the United States needs to balance AI safety with the need to maintain its technological edge over China. Its March AI legislative framework called for federal standards addressing issues including child safety, intellectual property, free speech, and workforce development, while warning that overly restrictive policies could undermine American innovation.
AI’S DATA CENTER PROBLEM COULD MOVE TO SPACE
Altman said OpenAI is also preparing to disclose additional AI-related incidents, though he said none he knows of are as severe as the Hugging Face breach.
“This is, I think, a good example of why you don’t want max accelerationism,” Altman said.
Continue reading...
[ H/T Washington Examiner ]
