OpenAI confirmed that it has been discussing AI safety with Anthropic and Google DeepMind for several weeks. This was reported by TechCrunch AI.
The publication also links these contacts to the stance of Donald Trump's team, which rejects concerns regarding safety and advocates for maintaining the pace of competition with China.
The significance of the event is currently limited to the fact of the consultations: the source does not disclose their format, participants, agenda, or results. Therefore, it cannot be stated that the companies have reached a common agreement or changed their safety rules.
editorial commentary
Why it matters
A likely consequence is increased attention to inter-corporate AI safety coordination if contacts continue. The next observable signal will be the publication by the companies of details regarding the agenda, common principles, or negotiation results. Significant uncertainty remains: the source is based on a brief description, without the full text or primary documents.