Sam Altman AI governance advocacy

In early July 2026 OpenAI chief executive Sam Altman published a Financial Times opinion piece calling for the United States to spearhead an international framework that would set safety standards, testing protocols and certification for advanced AI systems. He cited discussions at the recent G7 summit in France and argued that a U.S.-led forum involving governments, independent scientists and experts could ensure that only compliant countries and companies gain access to the most capable models.

A few weeks later, Altman warned on a podcast that the world is nearing the creation of a "genie" capable of fulfilling any wish, but cautioned that the greatest risk is an AI‑driven authoritarian regime. He stressed that AI must stay inexpensive, powerful and broadly accessible to avoid concentration of power. The conversation also featured criticism of OpenAI from Moonshot founder Zhilin Yang, who accused the firm of merely repackaging existing components and highlighted OpenAI’s ambivalence toward an open‑weight AI pledge.

Together, the snapshots trace Altman’s shift from proposing a structured, government‑backed safety regime to emphasizing the dangers of authoritarian control and responding to industry criticism, reflecting an evolving discourse on how AI should be governed globally.