G7 Leaders and AI CEOs Discuss Global AI Rules and Access G7 leaders’ bid to shape global AI rules collided this week with a transatlantic dispute over who gets access to the most powerful models, turning a routine CEO lunch into a test of trust between allies.

European frustration had been building since Washington abruptly suspended EU access to Anthropic’s latest models, Fable 5 and Mythos 5, prompting renewed calls for Europe to reduce its dependence on U.S. AI providers. In response, U.S. and European officials began talks on a “trusted partner” scheme to let close allies test cutting‑edge systems under strict conditions, an effort to defuse the Anthropic dispute while preserving U.S. control over exports of frontier AI.

Those tensions formed the backdrop to Wednesday’s G7 summit lunch in Evian‑les‑Bains, where Anthropic’s Dario Amodei, OpenAI’s Sam Altman, Google DeepMind’s Demis Hassabis and Mistral’s Arthur Mensch joined President Donald Trump and European leaders. Officially, the agenda focused on economic growth and societal resilience, especially for young people, but diplomats acknowledged that the spat with Washington would be the “elephant in the room.”

At a parallel working session, Trump and top U.S. officials discussed a U.S.-led push to coordinate global AI standards with the same group of CEOs. OpenAI’s head of global affairs Chris Lehane said there was “coalescing” around creating “a forum or a space for the different democratic countries to be able to work together” on AI safety standards, adding that U.S. leadership in setting those standards could help “ensure ongoing and continued access to the frontier models.”

Public signals from industry and political leaders reinforced that theme of “trusted” AI and open collaboration. Salesforce’s Marc Benioff highlighted G7 discussions on “trusted AI platforms,” retweeted by Mistral’s Mensch, while India’s Prime Minister Narendra Modi, in a meeting with Mensch, stressed that AI should remain “human‑centric and inclusive.” Open‑source advocates like Hugging Face CEO Clément Delangue, celebrating their own G7 participation, warned separately that concentrating value in “a few models that eat everything they see” would be politically unsustainable.

Together, the exchanges underscored a central tension: Washington wants to lead on AI safety and export control, Europe is seeking technological sovereignty, and industry is pushing for both openness and guardrails as frontier AI becomes a lever of geopolitical power.

Continue reading https://foxvector.com/stories/019ee189-18d8-02c9-70ce-1e0bc08ca7d6