OpenAI CEO Sam Altman reportedly laid out three possible endgames for AI safety in an internal presentation last week, according to a report from The Information cited by Techmeme and The Decoder. The remarks come as the Trump administration prepares its first framework for overseeing developers of superintelligence.
Altman described three paths: an international watchdog modeled after the IAEA, a merger of major AI labs into a single entity like CERN, or a scenario where one lab becomes so powerful through self-improving AI that it could shut down everyone else's compute. He reportedly called the last option 'terrible' and brought it up only for completeness, though one source said it was read as a jab at a competitor given AI safety theorist Eliezer Yudkowsky's ties to Anthropic.
The three AI endgames Altman described
The first approach would create an organization modeled after the International Atomic Energy Agency to enforce safety standards. OpenAI publicly called for such an institution back in May 2023. It would inspect systems, require audits, and monitor compute and energy use. A body like this would depend on international cooperation, especially with countries building frontier models, including China.
Altman called the second approach the 'Demis Plan,' after a proposal Demis Hassabis, then head of Google's AI lab DeepMind, made about ten years ago. Hassabis envisioned an AI organization modeled after CERN. In Altman's version, the major labs would merge into a single entity or operate under its umbrella, and that entity would supply cooperating countries with models.
Altman reportedly called the third approach the 'Eliezer Plan,' after AI safety theorist Eliezer Yudkowsky. In this scenario, a developer accumulates so much economic power through self-improving AI that it withholds compute from everyone else and could, in the most extreme case, shut down everyone else's computing capabilities entirely. Altman brought up the scenario only for completeness, according to the report. One source said the remark was read as a jab at a competitor, given Yudkowsky's ties to Anthropic.
