Generative artificial intelligence is beginning to reshape the Texas labor market, with employers pulling back on hiring for occupations whose work can increasingly be automated, according to new research from the Federal Reserve Bank of Dallas.
The analysis by Dallas Fed economists Samuel Dodini and Tucker Smith found that online job postings declined noticeably in occupations with greater exposure to generative AI after the release of ChatGPT in late 2022. The effect appears to be coming not simply from new, AI-oriented companies but from existing firms changing whom they hire.
The findings come as Texas businesses rapidly embrace the technology. Two-thirds of firms responding to the Dallas Fed’s May 2026 Texas Business Outlook Survey reported using AI, compared with 40% two years earlier.
To measure AI’s impact, the researchers used millions of online job postings collected by Lightcast and compared them with an occupation-level measure of AI exposure developed by Anthropic. That measure estimates the share of tasks within an occupation that generative AI can automate based on actual use of Anthropic’s Claude AI system.
Software development, web design and other computer-intensive occupations ranked among the most exposed. Managers, clerical workers and editors were also among white-collar occupations facing relatively high levels of exposure.
By the end of 2023, job postings for more AI-exposed occupations had declined about 5% relative to less-exposed positions. By the first quarter of 2025, the difference had grown to roughly 8%.
Existing companies showed a similar pattern. Firms that were more exposed to AI reduced their job postings about 5% to 6% by mid-2024 and 8% to 9% by early 2026. Those companies also changed the mix of jobs they advertised, shifting away from occupations containing more automatable tasks.
The Dallas Fed estimates that generative AI exposure reduced total Lightcast job postings in Texas by about 1.8% in 2024 and 2.6% in 2025. While relatively modest across the overall labor market, the researchers said the declines are more significant for workers seeking positions in highly exposed occupations.
The changes could fall particularly hard on younger workers. Fewer than half of the typical job postings in the Lightcast data require more than two years of experience, making online postings an important entry point for recent college graduates.
The researchers said administrative data from Texas indicate that AI automation has affected employment and earnings among recent college graduates and has prompted some current college students to alter their educational choices.
The authors cautioned that Lightcast does not capture the entire labor market. Jobs in farming, construction, building maintenance and personal services, among others, are underrepresented in online postings. Still, the findings offer early evidence that AI is not merely changing how Texans work. It is beginning to change which workers companies are looking to hire.