The scene at the French Alps this week would have seemed impossible just years ago. The CEOs of America's dominant artificial intelligence companies were seated around the table with presidents, prime ministers, and chancellors, treated not as guests but as peers with the power to shape global security and economic order.
President Trump sat flanked by OpenAI CEO Sam Altman on one side and Google DeepMind CEO Demis Hassabis on the other during the G7 working lunch in Évian-les-Bains. French President Emmanuel Macron, the summit's host, positioned himself alongside Anthropic CEO Dario Amodei and Salesforce CEO Marc Benioff. The message was unmistakable: these technology leaders now occupy a geopolitical tier formerly reserved for nation-states.
What happened behind closed doors reveals the central tension of our moment. The three largest AI company heads each made the same push to Western democracies: stay unified on AI governance, resist authoritarian competition, and establish international standards before unilateral decisions fracture the landscape.
Amodei warned the assembled leaders not to "splinter" as democracies roll out advanced AI capabilities. Altman called for an "international forum for discussion that establishes globally accepted standards for testing" and proposed that Western democracies cooperate on guardrails while preserving "human liberty." Hassabis invoked the scale of the moment itself, calling this "one of the most critical moments in human history" and framing AI advancement as potentially ushering in a new era of scientific discovery.
Altman also held bilateral meetings with multiple heads of state, the kind of one-on-one diplomatic sessions traditionally reserved for heads of government. The bilateral photographs followed protocol: AI CEO, world leader, flag backdrop. The symbolism was precise.
Yet Altman offered a notable caveat in his closed remarks. "Do not cede your responsibilities to AI labs like mine," he said. "No single lab should be making the decisions." The statement acknowledged what the visual composition contradicted: the outsized influence these private companies now wield over decisions that governments once controlled alone.
The working lunch included the CEOs of major AI labs from Japan, Germany, Italy, and the United Kingdom, as well as Arthur Mensch of the French company Mistral AI. The gathering crystallized a future that tech leaders and many government officials now openly discuss: constant negotiation between heads of state and the architects of AI systems over who controls development, who sets rules, and who bears responsibility when things go wrong.
This dynamic has already begun playing out in smaller skirmishes. Anthropic and other AI companies have clashed with the Trump administration over products deemed security threats. Those confrontations are essentially test cases for who ultimately decides how AI systems operate in the real world. The G7 moment suggests both sides are now preparing for a much larger negotiation.
Author James Rodriguez: "When tech CEOs need diplomatic protocols and bilateral meetings to negotiate their industry's future, you know the balance of power in the world has fundamentally shifted."
Comments