Enhance or Eliminate? How AI Will Likely Change These Jobs

Will generative AI replace your job or improve it? What has been the impact on the labor market so far?

After the public launch of ChatGPT in November 2022, job postings for occupations that involve lots of structured and repetitive tasks, likely replaceable by generative AI, decreased by 13%. Meanwhile, employer demand for jobs that require more analytical, technical, or creative work—potentially enhanced by artificial intelligence—grew 20%, according to a working paper coauthored by Harvard Business School Professor Suraj Srinivasan.

The findings offer early clues for how companies are adopting generative AI, which has sparked a corporate search for efficiency and existential dread among employees. The research team assessed job postings from 2019 through March 2025 using a large dataset that covers nearly all US vacancies.

“Rather than solely eliminating jobs, generative AI creates new demand in augmentation-prone roles, suggesting that human-AI collaboration is a key driver of labor market transformation,” says Srinivasan. The largest reductions were in the finance and technology sectors.

Srinivasan, the Philip J. Stomberg Professor of Business Administration, collaborated on the working paper “Displacement or Complementarity? The Labor Market Impact of Generative AI” with Wilbur Xinyuan Chen, of the Hong Kong University of Science and Technology, and Saleh Zakerinia of Ohio State University. The paper was first released in December 2024.

Occupations with potential for AI augmentation handle tasks that can be automated using generative AI alongside other tasks that require human involvement. Those most prone to augmentation tend to involve greater use of social and hands-on technical skills. Microbiologists, financial analysts, and clinical neuropsychologists are three examples with high augmentation potential. In finance, as Srinivasan explains, investment managers and analysts use AI-powered tools to process and evaluate market data, but ultimately, their judgment and decision-making remain crucial.

The research team used OpenAI’s ChatGPT to categorize over 19,000 job tasks across more than 900 occupations, assessing their potential for automation through generative AI. They also constructed an augmentation score based on the share of exposed and unexposed tasks in each occupation.

The number of skills required for roles prone to automation are shrinking, the researchers found. They registered 7% fewer of those skills in job postings and also fewer skills emerging in these occupations. At the same time, they detected more AI-related skills—such as prompt writing or using AI tools—in jobs with high augmentation potential. As workflows transform with the new technology, new skills have also emerged.

The researchers note that the study focuses on the short-term impact of generative AI on the US labor market, so the effects on other regions or long-term impacts “remain uncertain as adoption scales.”

How companies integrate generative AI technologies is decisive for job loss or growth, the paper warns. Given that it impacts jobs differently, Srinivasan recommends that companies:

• Invest in reskilling programs to transition workers to roles enhanced by AI. “Retraining is essential for jobs where generative AI is reducing skill diversity. In automation-prone occupations, workers may face displacement unless they develop non-automatable skills, such as judgment and interpersonal communication skills.”

• Continuous upskilling in generative AI to leverage new tools. “In augmentation-prone occupations, generative AI is broadening skill requirements, increasing the demand for AI literacy, human-AI collaboration, and domain-specific AI applications.”

“Firms should view generative AI as an augmentation tool rather than merely a cost-cutting measure and align workforce training programs accordingly to support both job transitions and evolving skill demands,” says Srinivasan.