At the G7 summit in France, Anthropic CEO Dario Amodei urged democratic nations to “resist the temptation to splinter” over artificial intelligence, calling for global unity. OpenAI CEO Sam Altman and Google DeepMind CEO Demis Hassabis supported the sentiment, with Altman stressing that advanced cyberdefense AI tools should be freely shared among friendly nations. The three executives called for a US-led collaboration and proposed creating a formal technical standards body and a US-led forum to evaluate model safety. The appeal came after the Trump administration blocked Anthropic from exporting its newest “Mythos” and “Fable” AI models to foreign markets over national security concerns, following reports that the models' safety guardrails could be easily bypassed. French AI startup Mistral's CEO, Arthur Mensch, noted that the fear of American software dependency dominated the roundtable, with non-U.S. participants repeatedly questioning whether counterparts could cut off access to an intertwined supply chain.