OpenAI, Anthropic, and Google DeepMind Started Writing the Safety Playbook Together

By Omar aka Crypt0 · Sep 17, 2026

The three labs competing hardest to build the future just admitted they have been meeting quietly for weeks about how to keep it safe. OpenAI's global policy chief Chris Lehane told Bloomberg that OpenAI, Anthropic, and Google DeepMind have been coordinating on AI safety, prioritizing it above the rivalry. The fiercest race in technology now has a shared pit crew.

The timing tells the story. Over the weekend, Anthropic CEO Dario Amodei published an essay titled We Must Pace the Frontier, calling on the industry to slow the rate at which it improves its most capable systems. OpenAI CEO Sam Altman promptly backed the call. Amodei asked the US government to mediate or at least enable the cooperation through a standards body or formal waivers, arguing that coordination among competitors carries antitrust exposure under the Sherman Act.

Washington is already building the legal scaffolding. In July, Senators Jim Banks of Indiana and Adam Schiff of California introduced the Collaboration on Adversarial Threats and Security Risks Act, which would give AI labs antitrust carveouts for safety coordination. Lehane, briefing reporters in Washington, said OpenAI sees such a waiver as unnecessary, pointing to precedents like airlines sharing safety information and Obama era guidance permitting companies to share information to defend against cybersecurity threats.

The legislative momentum keeps building. OpenAI is backing the FRONTIER Act, which would create a federal AI safety framework and require top labs to open their models to audits by independent verification organizations. A separate proposal taking shape under Majority Leader John Thune with Senators Ted Cruz and Amy Klobuchar would give the Commerce Secretary power to deploy government auditors to test AI models directly.

The pushback is substantive and worth hearing. Cohere CEO Aidan Gomez argues the cooperation lets market dominant labs write the rules for everyone else, calling the effort a wolf in sheep's clothing, a cartel by another name. FTC Chairman Andrew Ferguson says he is deeply suspicious of antitrust exemptions, seeing barriers to entry that would insulate incumbents from challenge. Meanwhile California signed SB 813 into law this month, creating a framework for state recognized independent verification organizations, building on last year's SB 53, which requires large frontier developers to publish safety frameworks and report critical incidents.

The honest read: rivals sharing safety notes is how mature industries behave. Airlines did it. Cybersecurity teams do it. AI is arriving at the same conclusion, with the disagreements happening in public, which is exactly where they belong.

What you are watching is the rulebook for the most powerful technology in history being drafted in daylight, with every objection on the record. Readers can follow the bills, the essays, and the audits as they develop. This is one of those rare moments when the future's governance is genuinely up for discussion.

Sources

New to crypto? Read the crypto glossary, browse frequent questions, read our story, or explore the story archive.