OpenAI confirmed that it is working with Anthropic and Google DeepMind on artificial intelligence safety measures. It is a rare effort by major AI rivals to coordinate on risks linked to advanced models. The talks have been underway for several weeks, OpenAI global policy chief Chris Lehane said at a Washington briefing on September 15.
Lehane said the companies are sharing information and discussing ways to strengthen safety practices. He also said the firms do not need an antitrust exemption for such cooperation. “It’s better to try to work together to prioritize safety,” Lehane said.
The discussions follow a 3,800-word essay from Anthropic CEO Dario Amodei that called for a slower pace of development for advanced AI systems. Amodei argued that researchers need more time to identify and address emerging risks as models become more capable. Reuters
OpenAI CEO Sam Altman backed the central idea of Amodei’s proposal. Elon Musk also expressed support for the call for greater caution in AI development. The debate brought safety concerns into sharper focus across the industry.
Lehane defended cooperation between competing technology companies by pointing to earlier examples of firms working together on safety. He compared the approach with cooperation seen in commercial aviation.
Lehane was also in Washington for discussions with US lawmakers on legislation covering severe risks linked to advanced technology. OpenAI indicated support for bipartisan proposals in the House and Senate.
The House proposal mentioned by Lehane was developed by Republican Representative Jay Obernolte and Democratic Representative Lori Trahan. Senate efforts include Majority Leader John Thune, Republican Ted Cruz and Democrat Amy Klobuchar.
The latest discussions show that AI safety is moving beyond internal company policies. Leading developers are now also exploring cooperation and legislative measures as AI capabilities advance.
Also Read: Who is Richer in 2026? Dario Amodei vs Sam Altman Net Worth Compared