OpenAI, Anthropic, Google have been in talks on AI safety for weeks
- ID
- 24716
- Status
- summarized
- Published
- 15 Sep 2026, 11:47 PM
- Fetched
- 16 Sep 2026, 12:37 AM
- Provider
- TechCrunch
- Category
- technology
- Original URL
- https://techcrunch.com/2026/09/15/openai-anthropic-google-have-been-in-talks-on-ai-safety-for-weeks/
- Source URL
- https://techcrunch.com/feed/
Summary
- Score
- 6.0
- Created
- 16 Sep 2026, 12:37 AM
- Tags
- Audience
- developersai_ml_learnerssaas_foundersai_agent_users
What happened
OpenAI's global policy chief Chris Lehane confirmed that OpenAI, Anthropic, and Google DeepMind have been in private AI safety talks for weeks, following Dario Amodei's essay calling for the industry to collectively slow frontier AI development. The three firms are reportedly working to create an industry standards body without U.S. government support, despite antitrust risk — Lehane said they don't need the government waiver Amodei proposed.
Why it matters
If these three labs form a self-regulatory standards body with the power to coordinate 'industry-wide slowdowns' (as Demis Hassabis proposed in July), it could directly affect which models you can access, how fast they ship, and what evaluation gates your AI products must pass. The antitrust question is unresolved and could derail or reshape whatever they agree on, so don't assume any announced standards will stick.
Discussion angle
Three labs coordinating safety standards without government backing — is this genuine risk reduction or a moat-building move that raises barriers for smaller builders and open-source models?