Regulation AI Minute Newsroom 2026-08-21

OpenAI has started a team whose stated worry is that a company like OpenAI ends up with too much power

On 20 August OpenAI published the first post of AI Futures, a blog and a newly named Strategic Futures team led by Dean Ball, the AI policy scholar who joined the company in June after a stint in Washington. The framing is unusual for a frontier lab. It argues that the largest and least-solved category of AI risk is not a model behaving badly but the concentration of power — a small number of companies, governments or institutions becoming disproportionately strong because of AI — and it sets itself the question of how a free society should be restructured to keep individual rights and agency intact as the technology arrives. Ball sketches guiding principles including a preference for individual responsibility over collective intervention and limits on how much visibility any single actor should have. It is a research and publishing effort rather than a governance commitment: nothing announced changes what OpenAI builds, who owns it, or how it is supervised.

Why it mattersRead one way this is a lab funding the study of its own risk, and the obvious objection writes itself — OpenAI is among the handful of organisations the concern is actually about. Read another way it marks where the argument is moving. For three years the public debate has been dominated by safety in the narrow sense, and the question of who ends up holding the power has been left largely to critics outside the industry. When a frontier lab makes it a staffed team with a name and a publication record, it becomes a subject legislators can quote back at that lab. The useful thing to watch is what the team publishes, not what it says it is for — that is the part that will be checkable against OpenAI's own conduct.

✓ Verified · 3 sources

Read in the app — free, in 9 languages

OpenAI has started a team whose stated worry is that a company like OpenAI ends up with too much power

Related stories