As artificial intelligence spreads across workplaces, researchers are trying to answer one of the biggest questions about the future of work: who is most vulnerable to automation.
A new analysis by researchers at GovAI and the Brookings Institution suggests the answer is more complicated than many predictions about mass job losses.
Their findings indicate that many workers whose jobs are highly exposed to AI may also be among the most capable of transitioning into new roles.
High Exposure Does Not Always Mean High Risk
The study examined more than 350 occupations, estimating how many tasks in each role could potentially be performed more efficiently using AI.
Jobs involving tasks such as computer programming, marketing, financial analysis, and customer service showed significant overlap with current AI capabilities. In theory, that makes them more exposed to automation, according to The Washington Post.
However, researchers also evaluated how easily workers could move into other jobs if AI changed or eliminated their roles. Factors such as education level, varied work experience, age, wealth, and access to strong job markets were used to estimate adaptability.
The results suggest many highly exposed workers also have skills that make career transitions easier.
Clerical Roles May Face Greater Challenges
Not all workers share the same flexibility.
The research identified about 6.1 million clerical and administrative workers as both highly exposed to AI and less able to adapt to new roles.
These jobs often involve repetitive tasks such as scheduling, document management, or coordination — activities that AI systems are increasingly capable of handling.
Researchers also found that women hold a large share of these vulnerable roles, accounting for roughly 86% of the workers most exposed to AI with limited adaptability.
Conflicting Evidence on AI and Jobs
Despite intense debate about AI’s impact on employment, economists say there is no clear evidence yet that the technology is reducing overall employment in the United States.
Studies examining AI’s effects on the labor market often reach different conclusions. Some research suggests AI adoption may already be affecting entry-level jobs in areas such as software development or customer service. Other analyses indicate workers in AI-exposed fields are performing better than peers in occupations less affected by automation.
Economists say the conflicting results reflect how early the technology still is in its adoption cycle.
Technology History Offers a Cautionary Lesson
Forecasting the labor impact of new technologies has historically been difficult.
Earlier predictions that automated teller machines would eliminate bank teller jobs or that early AI systems would replace radiologists failed to fully materialize. Instead, new roles often emerged alongside technological change.
Economists note that previous innovations—from electricity to the internet—eliminated some occupations while creating entirely new industries and job categories.
The Future of Work Remains Uncertain
Researchers emphasize that current forecasts about AI and jobs should be treated as estimates rather than definitive predictions.
While AI is likely to alter many occupations, the pace and scale of change remain unclear.
For now, the research suggests that while some workers—particularly in clerical roles—may face greater disruption, many others may have the skills and flexibility needed to adapt as the technology totally changes the workplace.
