Stronger AI, Fewer Jobs: The Half-Staff Economy Is Arriving Early
Companies in any market will apply the same arithmetic once AI is good enough: if half the staff can deliver the same output, the other half goes, and every capability improvement accelerates that decision. The practical consequence for Western developers is that job security based on today's AI limits is a short-lived assumption, and the time to reposition skills is now.
Each jump in AI capability pushes more companies to the same conclusion: they can halve the workforce without touching operations or revenue. The global economy was already weak, and AI's rapid evolution is compressing what would have been a gradual cycle shift into a sudden shock. No one can say whether the short-term pain lasts six months, two years, or longer.
The path forward, as argued, is that AI will eventually create new production relations and jobs, but not yet. Right now it only boosts productivity, and that's precisely the mechanism driving job cuts. Individuals are told to build core AI skills, prepare a financial cushion, and consciously let go of the industry expertise that once made them valuable, becoming generalists who work the way AI demands. Companies are told to make cost reduction and efficiency the top priority and rebuild workflows around AI immediately.
Many people still don't feel the risk, and that's fine because these changes take time. The window between early awareness and widespread adoption is the opportunity: what you do in that gap is up to you.
The claim that most companies can halve their workforce without operational impact is an unsupported assertion used as a rhetorical premise, not a measured finding.
Advising people to discard their proud industry experience contradicts the common view that domain knowledge compounds with AI; no evidence backs the stronger claim.
Framing the structural shock as a personal race to act within a 'time gap' shifts responsibility onto individuals and ignores aggregate consequences such as mass demand collapse.
Stronger AI forces more layoffs, which reduce consumption and profits, pushing firms further into AI adoption—a self-reinforcing spiral. On remedies, taxing AI and implementing universal basic income counters AI's unfair cost advantage over human workers, while a hands-on technical fix—local caching—halves API requests. Company headcounts have already dropped from hundreds to tens, though the future remains too uncertain to read any clear direction.
The stronger AI gets, the more unemployment, the less consumption, products pile up, profits decline, cost-cutting and efficiency improvements use AI.
Closed loop.
AI is actually engaging in unfair competition against ordinary workers. Businesses don't have to bear AI's wages, social security, or training costs, while AI's work capability clearly exceeds ordinary workers. In the future, we should move toward taxing AI and implementing UBI.
For AI, I first added a local cache layer, and cut requests by about half. It's a trade-off compared to the article's approach.