California deploys first AI job-loss tracker as white-collar cuts emerge
California has launched the first real-time dashboard to track AI-driven job losses, offering European policymakers a potential blueprint as early data points to white-collar disruption in the tech sector.
California has introduced the California AI-Unemployment Tracker (CAIT), a system that tags monthly unemployment claims with a score measuring how vulnerable a worker's previous role is to artificial intelligence. Governor Gavin Newsom unveiled the tool on Thursday, describing it as an early warning system built alongside the California Policy Lab and the state's employment department.
The initial data offers a nuanced picture that should alert European regulators and investors. While the state has not recorded a broad surge in AI-driven layoffs since the arrival of ChatGPT in late 2022, targeted pain is evident. Unemployment claims from college-educated workers in highly exposed roles have climbed and remained elevated through May 2026.
The geographic and sectoral breakdown is particularly telling for transatlantic markets. Claims rose sharply among AI-exposed workers in the San Francisco Bay Area and within technology-adjacent fields like information and professional services. There were no comparable spikes across demographics like race, gender, or age.
The tracker calculates exposure using two metrics: an academic measure developed with OpenAI assessing whether a model can perform at least half of a job's tasks, and Anthropic's index tracking actual Claude usage. Customer service representatives and software developers rank as highly exposed, while heavy-goods drivers and nursing assistants sit near the bottom.
For European governments grappling with the EU AI Act and looming labour disruptions, the Californian model is both a template and a cautionary tale. The methodology replaces guesswork with administrative data, but researchers acknowledge its blind spots. It misses gig workers and the self-employed, relies on unverified job titles, and cannot prove artificial intelligence caused any specific dismissal.
Despite these limits, the tracker sets a new standard for monitoring technological displacement. “This new tracker helps replace speculation with evidence,” said Till von Wachter, the UCLA economist who co-led the work, “giving us a clearer understanding of what’s changing.” As roughly one in five American workers now use AI tools, the pressure is mounting on both sides of the Atlantic to move from tracking disruption to actively mitigating it.
