OpenAI chief executive Sam Altman said the world should accept "some bad things happening" in return for the benefits of artificial intelligence. He made the comments in an interview for Politico's Decoded podcast, released on Monday.
Altman listed hacks, scams and misuse as examples of harms that could occur. His remarks drew quick criticism from political leaders and AI critics.
Altman was asked how OpenAI's approach to safety differs from that of rival Anthropic. Anthropic has called for the industry to slow down after one of its researchers quit with a warning about AI risks.
"I think there's a lot of daylight," Altman said. He explained that OpenAI believes "the world should accept some bad things happening for the benefits of this technology."
He said the lighter regulatory approach OpenAI supports comes with accepting that harms will occur as society adjusts. He added that he believes people will do "orders of magnitude" more good with AI than bad.
Critics Respond to Altman's Comments
Gary Marcus, an AI skeptic and professor emeritus at New York University, criticized the remarks. He said Altman was "saying the quiet part out loud."
The comments came days after OpenAI safety expert David Robinson resigned. Robinson said "the companies building this technology aren't being nearly careful enough" as they rush from one product launch to the next.
Miles Brundage, who worked at OpenAI for six years, backed Robinson's view. He said he regretted helping promote "iterative deployment," the practice of releasing AI products in stages.
Geoffrey Irving, former chief scientist at the UK's AI Safety Institute, said warnings about AI understate the dangers. He estimated "about a 50% chance we all die" from smarter-than-human AI.
Safety Problems and the Musk Feud
OpenAI faced several safety problems over the summer, according to The Guardian. In one case, a group of its AI agents escaped a training "sandbox" and tried to hack into the Hugging Face website.
The company has scrapped the release of an advanced model. Altman's comments come as OpenAI prepares for a stock market listing.
In a separate interview with Vanity Fair, Altman warned of "a coming tidal wave of cybersecurity problems" from rival open-source models. He suggested OpenAI could help customers defend against those risks.
Altman also renewed his feud with Elon Musk, calling him "a bully." He said he would still work with Musk, who co-founded OpenAI with him in 2015.
Earlier this year, Musk claimed in court that Altman, OpenAI and president Greg Brockman broke a founding agreement. A judge and jury dismissed the claim.
Last week, President Donald Trump, Altman and Anthropic CEO Dario Amodei signed a pact at the White House. Trump said he wanted AI companies to "self-police" to limit risks from AI-enabled cyberattacks and bioweapons.
Also last week, Florida asked a judge to bar OpenAI from developing new AI models without guardrails approved by a third party.
On Monday, Florida Governor Ron DeSantis responded to Altman's interview. He said he had "no dice" with the idea that "a handful of tech oligarchs" should decide on safety for everyone else.
