California's AI jobs tracker detects no automation-driven unemployment in its first data

By CBS News · June 26, 2026.

California becomes the first U.S. state to have a public dashboard designed specifically to track employment trends related to artificial intelligence: both the jobs created and those destroyed by automation. The initiative, the first of its kind nationwide, responds to growing political and social pressure to measure with real data what until now had been, to a large extent, a speculative discussion.

According to the first findings published by the tool, there is no evidence of an increase in state unemployment attributable to jobs exposed to AI. This result does not necessarily imply that automation is not having effects, but rather that, at least in this early phase and with the available data, the Californian labor market does not register a statistically significant negative signal.

Till von Wachter, faculty director of the California Policy Lab at the University of California, Los Angeles (UCLA), is one of the academics responsible for analyzing and interpreting the tracker's data. Von Wachter explained the findings on CBS News' 'The Takeout' program, although the detailed content of his remarks is not available in the article's text, as it is a video.

As sector context, the debate over AI's impact on employment has divided economists and technologists for years. Previous studies —such as those by the McKinsey Global Institute or the World Economic Forum— have projected both massive job displacement and the creation of new roles, with final outcomes highly dependent on the pace of adoption and on workforce retraining policies. The creation of an official real-time tracking instrument represents a relevant methodological step: moving from projections to empirical observation.

The fact that California is the first state to implement this type of tool is no coincidence. The state is home to the world's largest concentration of AI companies —from OpenAI to Google DeepMind, Anthropic, and Meta AI— and its legislature has been particularly active in attempting to regulate the technology, albeit with mixed results. Having its own data on the labor market strengthens the state's ability to make evidence-based public policy decisions.

The risks of interpreting these first data points are evident: the effects of technological displacement tend to appear with a lag, and an early snapshot can be misleading. In general, economists warn that generative AI is being adopted primarily as a productivity tool —boosting the capacity of existing workers— before directly replacing them, which could explain the absence of a signal in aggregate unemployment at this stage.

For companies and developers of agentic AI tools, the launch of this tracker has direct implications: if the data confirms over time that AI does not generate net unemployment, the regulatory 'emergency brake' argument loses force. If, on the contrary, the data begins to show displacement, California will have the evidence needed to justify more aggressive legislative interventions.

From a regulatory perspective, this type of monitoring infrastructure parallels what the European AI Act requires regarding transparency and impact assessment. The difference is that California is implementing it proactively, before an equivalent federal legal obligation exists in the United States.

Looking ahead, the tracker's real value will depend on its granularity —whether it distinguishes by sector, education level, type of task, or region— and on how often it is updated. An instrument that only offers aggregate data at the state level may conceal severe impacts on specific groups or industries, even if total unemployment remains stable.