AI will eliminate more jobs than it creates

The framing is wrong. Every general-purpose technology in the last 250 years — the steam engine, electricity, the personal computer, the internet — was met with the same prediction: it will destroy more jobs than it creates. The prediction has been wrong every time, not by a little, but by orders of magnitude.

What AI does is collapse the cost of cognitive tasks the same way industrial machines collapsed the cost of physical ones. Cheaper cognition expands the demand for everything cognition unlocks — analysis, design, coordination, judgement at scale. The historical pattern is that whole new categories of work appear (data analysts didn't exist in 1990; software engineers were exotic in 1970; pilots, radiologists, and YouTubers each took an existing capability and built a profession around it).

The real risk isn't elimination. It's that the transition is fast enough to outpace re-skilling for individual workers. That's a policy problem, not a technology problem — and it's the conversation you should be having instead.