Understand this story
According to a report by The Information citing people familiar with the matter, OpenAI CEO Sam Altman last week internally presented employees with three paths for AI safety: first, establishing a regulatory body modeled on the International Atomic Energy Agency (IAEA); second, a lab merger proposal known as the Demis Plan; and third, a single-developer compute monopoly proposal known as the Eliezer Plan. The report also said Altman complained that there are too few proposals for slowing down AI development, while stating that OpenAI cannot afford to fall behind in the competition, and that he plans to decide on an IPO path within a year. The report mentioned that the United States currently relies on voluntary pre-review of new models, and in June imposed export controls on Anthropic on the grounds that the Fable model posed a jailbreak risk. At present, this claim comes only from the single report above, and there have been no subsequent developments with supporting evidence.
AI-generated from public reports
Latest development · Report
Sam Altman Reportedly Presented Three AI Safety Endgame Options to EmployeesDevelopments
1 confirmed developments
- Source:The Decoder
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 21
The Decoder
Sam Altman Reportedly Presented Three AI Safety Endgame Options to Employees
The Information, citing people familiar with the matter, reported that OpenAI CEO Sam Altman last week internally proposed three AI safety paths: establishing a regulatory body modeled on the International Atomic Energy Agency (IAEA), a lab merger plan known as the Demis Plan, and a single-developer compute monopoly plan known as the Eliezer Plan. Altman also complained that there are too few speed-limit proposals for AI development, while noting that OpenAI cannot afford to fall behind in the competition, and plans to decide on an IPO path within a year. The United States currently relies on voluntary pre-review of new models, and in June imposed export controls on Anthropic on the grounds that the Fable model posed a jailbreak risk.