In the autumn of 2026, Nvidia — the company whose chips have become the sinew of modern artificial intelligence — is in talks to anchor Anthropic's public offering with as much as ten billion dollars. The potential deal is less a single transaction than a symbol: of how swiftly AI has moved from research curiosity to the organizing force of capital markets, and of how deeply the largest technology firms now feel compelled to shape who builds the future. Anthropic, born from a schism at OpenAI and devoted to the question of whether powerful AI can be made safe, would emerge from such a deal not