The Kremlin said it does not rule out resuming trilateral talks in October.

Anthropic CEO Dario Amodei said in a long blog post the firm will adopt new safety measures, including third‑party assessments, and urged an industry‑wide slowdown in AI model development. OpenAI CEO Sam Altman said he will adopt Amodei’s proposal to grant independent assessors employee‑like access. xAI’s Elon Musk publicly backed Amodei. Despite longstanding warnings from these leaders, coordinated restraint would be unprecedented in a fiercely competitive sector where firms regularly launch new products to boost engagement and sales. It remains unclear how extensively leading AI companies will implement new limits or safety checks.

The Wall Street Journal reports Prime Minister Carney proposed making Canada the EU’s first "quasi-member state." European and Canadian officials say Brussels is receptive; US officials are reported to be skeptical.