"AI replacing human workers" — news sentiment
March 1 – July 19, 2026 · every jobs/workforce-relevant article from the five monthly news sheets, classified as alarming, reassuring, or neutral
Net sentiment by week
Net sentiment = (reassuring − alarming) ÷ all relevant articles that week. −1 means every article was alarming; +1 means every article was reassuring. Hover a point for article volume.
Coverage volume by week
Neutral/analytical articles are counted in the totals and tooltips but not drawn as bars.
Month by month
Each month links to its own page with a daily breakdown and a forecast written from that month's coverage alone.
Current forecast (as of July 19)
The short version: coverage has been net-alarming for five straight months, and after genuinely softening through May and June, it is darkening again. No month has come close to net-positive.
March (−0.41) was the deepest raw wave: companies openly naming AI as the reason for mass cuts — Block (40% of staff), Atlassian, HSBC, Crypto.com, recurring Meta reports — with trackers counting 45–59K cuts and roughly 1 in 5 explicitly AI-linked. The counter-narrative ("AI-washing" — blaming AI for cuts it didn't cause) was born here too, voiced by economists, Nvidia's CEO, and Block's own workers.
April (−0.38, thin data) shifted from events to dread: Goldman's "world of pain" for displaced workers, a "job-pocalypse" forecast, and the first worker-resistance stories (Gen Z quietly sabotaging AI rollouts). Note the April sheet mostly contains May-dated rows, so this month is undersampled.
May (−0.15) was the closest to balanced the narrative ever got. The layoffs kept coming (Meta ~8,000, Cloudflare ~20%, Cognizant 15,000, AI named the top layoff cause two months running), but the pushback peaked: Altman calling out AI-washing, Gartner finding no ROI behind AI-justified cuts, a steady "AI transforms, doesn't displace" drumbeat from the cybersecurity field, strong skills-premium data, and the first policy response (Newsom ordering the state to ease AI-layoff pain).
June (−0.19) turned the fight onto entry-level jobs: 1-in-3 employers admitting they replace junior roles with AI, Goldman raising its displacement forecast, and a layoff tally climbing past 158K — against an unusually vocal, executive-led defense (AWS hiring 11,000 juniors and calling junior displacement "bad for business", a Google exec blaming copycat layoffs, data showing heavy AI adopters actually grew headcount).
July so far (−0.38) has undone most of that recovery. The month opened with real green shoots — job cuts down 53% in June, companies rehiring workers after failed AI replacements — but the Meta lawsuit (AI allegedly selecting workers on medical and maternity leave for layoffs) changed the story's register: AI is no longer just the stated cause of layoffs, it's accused of being the mechanism that decides who goes. Add Google workers petitioning for layoff protections, a tracker claiming AI now drives over half of 2026's cuts, and the month closing on "the era of the forever layoff."
Where it points: the push-and-pull is real — the skills-premium and AI-washing-skepticism threads have never disappeared, and they produced genuine rebounds in May and June. But each rebound has been weaker than the negative wave that followed it, and the nature of the negative coverage is escalating: from layoff counts (March) → displacement fear (April) → contested causation (May–June) → legal and accountability fights (July). Expect August coverage to stay net-negative, driven less by layoff tallies and more by lawsuits, worker organizing, and regulation — the Meta case, the Google petition, and Newsom's order are the template.