Chris Lehane, OpenAI’s global policy chief, confirmed on Tuesday that the company has been coordinating with rivals Anthropic and Google DeepMind on AI safety for weeks. Lehane was in Washington to meet U.S. lawmakers regarding catastrophic risks from artificial intelligence.
The push for cooperation
Reports emerged after Anthropic CEO Dario Amodei published an essay on Saturday. He argued the industry must slow the pace of frontier AI development to prevent disaster. Leaders including OpenAI’s Sam Altman, Google’s Demis Hassabis, and SpaceXAI’s Elon Musk backed Amodei’s letter. Altman stated OpenAI would join Anthropic in hiring third-party evaluators to monitor safety.
Lehane admitted the three firms have discussed safety matters for several weeks. Altman and others warned such coordination could violate antitrust laws if it suppresses competition. Amodei suggested a narrow government waiver was needed, but Lehane reportedly said a waiver is unnecessary.
Behind the scenes
The news confirms earlier reporting that the companies were working together. In an interview with Fortune late last week, Altman hinted at private discussions with other AI leaders. This week, The Information reported the three firms are building a standards body for the AI industry. Altman told staff this effort must happen without U.S. government support.
Hassabis called on the U.S. in July to establish a new standards body to act as an AI watchdog. This body would screen the world’s most advanced models and coordinate industry-wide slowdowns if dangers progressed.
Political resistance
President Trump has dismissed safety concerns as a hoax and argued against tighter regulations. He stated any slowdown would give China an advantage in the global race. His key AI advisor David Sacks, a longtime investor with stakes in much of the industry, echoed this view. Sacks said fears of existential risk from AI are overblown.
At the Tuesday meeting, Lehane said OpenAI supports a provision in the FRONTIER Act. This measure would force top frontier labs to allow independent verification organizations into their companies to ensure models are developed safely.
TechCrunch has reached out to Anthropic, Google, and OpenAI for comment.
What it means
For people making things, the situation changes how tools are built and released. Independent evaluators will have access to labs, meaning developers cannot hide model behaviour. Industry slowdowns could delay new features or capabilities if safety risks are identified. The private standards body suggests rules might come from companies rather than regulators, creating a system where the builders decide the speed of innovation.
