AI CEOs Unite at G7 to Push US-Led AI Coalition
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Dario Amodei, Sam Altman, and Demis Hassabis attended a G7 CEO lunch in France on June 17, 2026, publicly calling — unusually together — for US-led international collaboration on AI rules rather than fragmented national approaches
- Trump told reporters negotiations with Anthropic are "going fine" even as a US-driven shutdown of Anthropic's models drags on, and reportedly told leaders "I'm the boss" at the summit (WSJ, Reuters, Newsweek)
- Macron is actively seeking a workaround to Trump's block on Anthropic's AI models while simultaneously "rolling out the red carpet" for Trump after months of geopolitical friction (Bloomberg Law, Courthouse News)
- Amodei publicly explained his split from OpenAI by saying he left because he doesn't trust Altman — a rare on-record jab at a rival CEO at a moment when the two are sharing a table (Business Insider)
- Mark Cuban broke from the G7 unity picture, calling Altman "all over the map" and warning that "at some point, people stop trusting" him, undercutting the coalition-building narrative (Barchart)
- Altman used the same G7 stage to tell leaders not to "cede responsibilities" to AI giants — a noticeably different posture from Amodei's resistance framing, and one European leaders seized on to argue for sovereignty checks (Economic Times, AP)
- The underlying tension TechCrunch flagged: world leaders want access to American AI but do not want the US to retain a kill switch — the unresolved fault line the summit exposed rather than resolved
Why it matters: Three rival AI CEOs sharing a G7 lunch table is itself a signal of where power now sits in tech diplomacy, but the fractures underneath — Amodei openly distrusting Altman, Cuban attacking Altman by name, Macron maneuvering around a US ban, and leaders demanding kill-switch-proof models — show the "US-led AI coalition" is a slogan, not a deal. The 77-day US shutdown of Anthropic's models is the live policy pressure point; any G7 output that gives European or allied governments a way to access Anthropic without US veto authority directly threatens the US leverage Trump is defending.
