A360
Key Takeaways✦ Atlas AI
01
Nearly one-third of American workers now frequently use AI, but this is matched by rising anxiety about job displacement and significant skepticism from non-users.
02
AI users, particularly in managerial roles, report notable productivity gains, with two-thirds seeing a positive impact on their efficiency at AI-adopting companies.
03
Resistance to AI is rooted in a preference for existing methods, ethical objections, data privacy fears, and concerns about the technology's reliability and accuracy.
Atlas AI
U.S. workplace AI adoption is increasing, but a new Gallup survey shows employees are split between seeing productivity gains and worrying about job security. The February poll found that about three in ten American employees now use AI tools frequently, defined as daily or several times a week. At the same time, the survey points to growing concern that new technologies could make some roles obsolete.
Gallup’s findings suggest the divide is not only about access to tools, but also about how workers experience them. Around four in ten workers said their company has integrated AI technologies. Within that group, roughly two-thirds reported that AI has improved their personal productivity and efficiency, indicating that many users associate the tools with faster or smoother workflows.
The reported benefits vary by seniority and job type. About 70% of leaders who use AI at least a few times a year said it increased their efficiency, compared with just over half of individual contributors. The survey also indicates that workers in management, healthcare, and technology are more likely to report productivity improvements than those in service roles, where 45% said they saw similar gains.
Gallup’s results also highlight how AI is being applied in targeted ways rather than across every task. The survey cited examples of professionals using generative AI to draft diplomatic communications in adversarial fields, illustrating how some users deploy the technology for specific, non-core work that can be standardized or accelerated. This pattern aligns with the broader theme in the poll: adoption can be meaningful for certain functions, while remaining limited or contested elsewhere.
Despite wider availability, a large share of employees remain infrequent users. Gallup found that half of all U.S. employees use AI tools once a year or not at all. Among employees who have access to AI but do not use it, 46% said they prefer their existing work methods, suggesting inertia or satisfaction with current processes can be as important as technical capability.
Other barriers are tied to trust and governance. Another 40% of non-users with access cited ethical opposition, data privacy concerns, or a view that AI is not useful for their job. Reliability concerns also featured prominently, with some professionals pointing to “hallucinations,” where AI generates false information, making certain tools unsuitable for high-stakes work such as legal research where errors can carry serious professional consequences.
Finally, Gallup reported that about one-fifth of non-users with access said they do not feel adequately prepared to use AI effectively, pointing to a skills gap that could slow implementation. The survey leaves open key uncertainties for employers and policymakers, including whether training, clearer rules on data use, and improved tool accuracy will shift adoption patterns—or whether resistance will persist even as AI becomes more embedded in workplace systems.
