Altman Warns AI Risks of Lost Control and Power Concentration

OpenAI CEO Sam Altman has publicly warned that artificial intelligence development poses serious risks, including the potential loss of human control and dangerous concentrations of power. In a series of statements and interviews, Altman outlined two primary scenarios where AI progress could go "very badly," urging the industry to pace its development to allow safety measures to keep pace. Altman detailed the risks in a social media post on September 14, 2026. He stated the first danger is that humans could "lose control of the future to AI," which he called unacceptable. The second risk is the concentration of too much power in one person, company, or country, which could lead to an "extremely dystopian" outcome if an extraordinarily powerful AI is used to impose a single worldview on everyone else. These comments followed Altman's endorsement of a proposal by Anthropic CEO Dario Amodei to slow the rate of frontier AI model development. Altman agreed with Amodei's call for pacing and committed OpenAI to providing independent evaluators with employee-like access to its systems to verify safety practices.