The AI doom scenario you've never heard of may be the hardest to stop

Why it matters

Researchers are sounding the alarm about 'gradual disempowerment,' a scenario in which AI steadily displaces human labor, replaces human decision-making, erodes creativity, and degrades cognition. The term originates from a paper published in early 2025. Jay Chooi, founder and CEO of Robocurve, a company that tracks AI-driven robotics capabilities, ranks it at the top of his risk list, comparing it to engineered pandemics and destructive cyberattacks. He calls it a civilization-scale problem requiring a whole-of-society effort. The article arrives amid near-daily releases of AI assistants, including Muse by Meta, Gemini by Google, and dots by OpenAI, marketed as tools that take tasks off users' plates. Critics warn that as humans delegate more work and thinking to machines, they risk becoming passive 'approvers' whose skills and oversight capacity deteriorate. The piece illustrates the dynamic with a hypothetical manufacturer forced by market pressure to replace its workforce with robots, and cites AI research scientist Margaret Mitchell, who argues that people who stop practicing skills become reliant on systems no human can fix. The concern extends beyond jobs to democratic governance and military decision-making.