An analysis by Dallas Fed economists finds that generative AI exposure is beginning to reshape Texas hiring. The researchers estimate AI automation exposure reduced Lightcast online job postings by 1.8% in 2024 and 2.6% in 2025. The findings show where hiring pressure is emerging and which workers may feel it first.
Key Takeaways
- Two-thirds of Texas businesses surveyed in May 2026 reported using artificial intelligence, compared with about 40% two years earlier.
- Job postings fell more sharply in occupations whose tasks were more exposed to generative AI automation.
- The researchers estimate AI automation exposure reduced total Lightcast job postings in Texas by approximately 1.8% in 2024 and 2.6% in 2025.
- The analysis measures hiring demand through job postings. It does not show that AI caused an equivalent decline in total Texas employment.
- Recent graduates and other people entering or changing careers may face greater pressure because online postings tend to represent jobs requiring relatively little prior experience.
Dallas Fed Analysis Finds AI Pressure in Texas Hiring
Generative artificial intelligence is beginning to leave a measurable mark on the Texas AI job market, according to research published Sept. 1 by Federal Reserve Bank of Dallas economists Samuel Dodini and Tucker Smith.
Their analysis found that after ChatGPT was released in November 2022, Texas employers reduced job postings for occupations with tasks that generative AI could automate relative to less-exposed occupations. By the end of 2023, postings for more-exposed positions had fallen about 5% relative to less-exposed roles. The difference reached roughly 8% by the first quarter of 2025.
The researchers used millions of Lightcast online job postings to examine hiring activity by occupation and industry. They paired those records with a task-based measure of AI automation exposure derived from actual use of Anthropic’s Claude platform and O*NET occupational task classifications.
The approach allowed the researchers to compare hiring patterns among occupations with different levels of exposure while accounting for industry differences.
That distinction matters. The results do not show that artificial intelligence caused every decline in Texas job postings or that existing workers in AI-exposed occupations were necessarily laid off. Instead, they identify a relationship between greater exposure to AI automation and weaker demand for new hires.
Job Postings Fell More in AI-Exposed Occupations
The difference became increasingly visible as businesses adopted generative AI.
The researchers found that existing firms with greater exposure to AI reduced postings by roughly 5% to 6% by mid-2024 and by 8% to 9% by early 2026. The results suggest the hiring shift was not driven primarily by a wave of new AI-focused companies or by the disappearance of older businesses.
Instead, established employers also appeared to change the types of workers they were seeking.
Companies that had been more exposed to automation before ChatGPT’s release subsequently advertised fewer positions involving tasks that could be automated. That change in the composition of postings offers evidence that employers were adjusting hiring demand as generative AI capabilities expanded.
Dodini and Smith wrote that there is “strong evidence that GenAI has decreased labor demand” for occupations involving tasks the tools can perform.
The impact remained uneven. Occupations with greater automation exposure experienced larger reductions in advertised openings than jobs whose tasks were less suited to generative AI.
AI Exposure Was Associated With a 2.6% Posting Reduction
The researchers also estimated how occupational AI exposure affected overall Lightcast posting activity in Texas.
Their calculations indicate that exposure to generative AI automation reduced total Lightcast job postings in Texas by approximately 1.8% in 2024 and 2.6% in 2025.
Those figures represent estimated effects associated with AI automation exposure. They should not be interpreted as evidence that artificial intelligence was responsible for every change in the state’s labor market during those years.
Texas companies continued expanding and creating jobs in other areas. A recent San Antonio manufacturing expansion, for example, involves planned hiring across manufacturing, engineering, logistics and corporate support. Such projects illustrate why changes in AI-exposed hiring should be separated from the direction of the overall Texas labor market.
The Dallas Fed’s Aug. 21 employment forecast projected that Texas would add about 241,300 jobs during 2026, representing 1.7% annual job growth. That broader outlook can coexist with weaker demand in specific occupations whose tasks are increasingly exposed to automation.
Texas Businesses Are Adopting AI at a Faster Rate
The hiring findings come as artificial intelligence use has spread rapidly among Texas businesses.
Two-thirds of firms responding to the May 2026 Texas Business Outlook Surveys said they were currently using AI. The survey collected responses from 313 Texas business executives between May 12 and May 20.
That adoption rate was unchanged from December 2025 but higher than 59% a year earlier. Dallas Fed research also shows that roughly 40% of responding businesses were using AI in 2024.
The May survey found that AI adoption remained more common among service-sector businesses than manufacturers. Much of the reported use was still concentrated in testing, pilot programs or deployment among relatively small groups of employees.
The findings therefore describe a technology transition that is still developing rather than a uniform transformation across every employer.
Anthropic’s own Economic Index research similarly shows that AI use varies substantially across occupations and individual job tasks. Its research distinguishes between automation, where an AI system performs a task more directly, and augmentation, where people use AI collaboratively.
Early-Career Workers Face Greater Exposure to Hiring Changes
The decline in advertised openings could matter most for people trying to enter the labor market.
The Dallas Fed analysis notes that fewer than half of typical job advertisements in the Lightcast data explicitly require more than two years of experience, while relatively few require more than five years. That makes online postings especially important for recent graduates and workers attempting to move into new occupations.
A reduction in openings can therefore affect job seekers even without widespread layoffs among existing employees.
Separate New York Fed data reinforce the difficult environment facing younger college graduates nationally. In the second quarter of 2026, unemployment among recent college graduates remained around 5.6%, while the underemployment rate was approximately 42%.
Texas is also investing in pathways into occupations where employers continue to need specialized workers. A recent Texas aerospace workforce initiative connects employers with community colleges, career and technical education programs and workforce organizations to develop pipelines into skilled aerospace positions.
Together, those developments show that the effect of AI is unlikely to be uniform across occupations. Hiring pressure may intensify in jobs whose tasks can be automated while demand persists or grows in technical, industrial and other specialized fields.
Job Postings Measure Hiring Demand, Not Total Employment
The Dallas Fed findings are most useful when interpreted within the limits of the data.
Online job postings provide a timely measure of employer hiring intentions, but they do not capture every available position or every change in employment. A company can reduce new hiring without laying off existing employees, while some jobs may be filled through channels that do not appear in online posting databases.
The analysis therefore provides evidence of changing labor demand rather than a complete count of jobs eliminated by artificial intelligence.
It also does not establish that every occupation with high AI exposure will experience the same outcome. Jobs consist of different combinations of tasks, and artificial intelligence may automate some activities while assisting workers with others.
That distinction is also reflected in Anthropic’s research, which finds both automation and augmentation in real-world Claude use rather than a single pattern of AI replacing human work.
The Texas AI Job Market Is Becoming More Uneven
The emerging picture of the Texas AI job market is not one of uniform job losses. Instead, the data show a widening difference in hiring demand depending on how much of an occupation’s work can be performed by generative AI.
AI adoption among Texas firms has increased substantially since 2024, while occupations with greater automation exposure have recorded larger declines in online postings. The researchers’ estimate of a 2.6% reduction in total Lightcast postings in 2025 suggests that the statewide effect remains modest compared with the sharper changes experienced by particular categories of workers.
For employers, workers and educators, the significance lies in the composition of hiring rather than a single statewide employment number. Texas can continue adding jobs overall while opportunities shift away from some occupations and toward others.
The next phase of the Texas AI job market will depend not only on how quickly businesses adopt new tools, but also on which tasks employers automate, which skills they continue to hire for and how workers entering the labor market adapt to those changes.
Frequently Asked Questions
What does the Dallas Fed analysis say about AI and Texas jobs?
The analysis finds that occupations with greater exposure to generative AI automation experienced larger declines in online job postings than less-exposed occupations. The researchers estimate automation exposure reduced total Lightcast postings in Texas by about 1.8% in 2024 and 2.6% in 2025.
How much has AI adoption increased among Texas businesses?
Two-thirds of firms responding to the May 2026 Texas Business Outlook Surveys said they were using AI. Dallas Fed data indicate the comparable share was about 40% in 2024.
Which jobs are most affected by AI in Texas?
The Texas AI job market data show larger posting declines in occupations whose tasks have greater exposure to generative AI automation. The analysis evaluates task exposure rather than identifying every occupation as either affected or unaffected.
Does the research show that AI is causing layoffs in Texas?
No. The study primarily measures changes in online job postings and therefore provides evidence about hiring demand. A decline in postings does not establish an equivalent decline in existing employment or prove that workers were laid off because of AI.
Why could recent graduates be more affected?
Many online job postings require relatively little previous experience, making them an important entry point for recent graduates and career changers. A decline in those opportunities can make entering or switching occupations more difficult even without widespread job losses among existing employees.
AI hiring trends Texas, Texas job postings, AI automation jobs, Texas workforce data, generative AI employment
