OpenAI, Anthropic and Google secretly joined forces to collaborate on AI safety

OpenAI Group PBC’s global head of policy Chris Lehane said today that his company has been working closely with rivals Anthropic PBC and Google LLC’s DeepMind research lab to try to devise a framework that ensures artificial intelligence safety.

“It’s better to try to work together to prioritize safety,” Lehane said at a press conference in ​Washington, according ⁠to a report by Bloomberg. His comments came in the wake of a number of AI industry leaders sounding the alarm bell about the trajectory and pace of frontier model development.

On Saturday, Anthropic Chief Executive Dario Amodei published an essay on his personal blog, calling for the AI industry to cooperate in order to try and slow down the rapid rate of frontier AI’s progress in order to avoid what he termed “catastrophic risks.” Surprisingly, Amodei’s call won the support of the leaders of all of his company’s major rivals, including OpenAI CEO Sam Altman (pictured), SpaceXAI Corp.’s Elon Musk and Google’s Demis Hassabis. Altman even said OpenAI is willing to join Anthropic’s initiative to start embedding third-party evaluators within its development teams to monitor AI safety.

Today, Lehane told reporters in a media briefing that his company has been working with Anthropic and Google on AI safety initiatives for several weeks. He didn’t elaborate on the nature of their work, but some, including Altman, have raised concerns that the companies could risk ending up in violation of U.S. antitrust laws if their collaboration is perceived as harmful to competitors.

For his part, Amodei said in his essay that the U.S. government should consider giving the companies a waiver to enable safety cooperation without falling foul of antitrust laws, but Lehane said today that he thinks such a move is unnecessary.

Some legal experts have warned that any coordination on safety by the AI leaders could be construed as “output restriction,” which would mean they violate antitrust laws under the Sherman Act. These fears were previously raised by Indiana Sen. Jim Banks and California Sen. Adam Schiff, who proposed a “Collaboration on Adversarial Threats and Security Risks Act” to enable AI labs to act on safety, even if those actions contravene antitrust law.

Any efforts by the companies to try to secure some kind of waiver are likely to face stiff opposition. In a blog post published in response to Amodei’s essay, Cohere Inc. CEO Aidan Gomez said the safety calls look like an effort by the top AI labs to disguise an anti-competitive scheme that would preserve their lead in frontier model development. He said it would result in “market-dominant” labs being allowed to “define the rules and safety standards” for the rest of the industry.

“Once again using fear under the pretext of protecting the public, these oligopolies are now requesting to bend competition rules and be permitted to dictate the terms for everyone else,” Gomez said. “A wolf in sheep’s clothing, a cartel by another name.”

Gomez may be a much smaller voice, but he potentially has a very powerful ally. U.S. President Donald Trump has consistently pushed back against AI safety concerns, dismissing them as a hoax and rejecting calls for tighter regulation. In a post on Truth Social, Trump rejected Amodei’s call, arguing that any slowdown would enable China to overtake the U.S. in the AI race.

He’s backed up by his key AI adviser David Sacks, who is a longtime investor with stakes in many AI firms. He has gone on record to say that fears AI could present an existential threat are overblown.

Lehane’s comments came during a media briefing in Washington, where he was due to meet a number of lawmakers. The company is reportedly throwing its weight behind two bipartisan bills affecting the AI industry. The first is the FRONTIER Act, which calls for the creation of a federal framework for AI safety and mandates that AI labs submit their new models for audits by independent evaluators.

OpenAI is also believed to back a second proposal that’s currently under development. According to Reuters, Majority Leader John Thune is working with Republican Sen. Ted Cruz and Democratic Sen. Amy Klobuchar on a new bill that would give the Commerce Secretary the power to subject new models to safety audits prior to their release.

Photo: World Economic Forum Annual Meeting/Flickr