For the first time, 50% of employed US adults say they use AI at work at least a few times per year; leaders are most likely to see AI's impact as positive
Employees report productivity gains with AI but not fundamental shifts in how work gets done. — For the first time in Gallup's measurement …
Context & Ripple Effects
Workplace AI use has risen steadily in Gallup's series: employees using it at least occasionally climbed from 21% in 2023 to 40% in 2025, while more frequent use also increased in the prior Gallup workplace-adoption survey. The new threshold suggests occasional AI use is becoming a normal part of many roles rather than a niche practice.
The pattern remains uneven: Gallup had found only 12% using AI daily in late 2025 and nearly half reporting no use at all in the most recent daily-use reading. Productivity gains without fundamental workflow change indicate adoption is broadening faster than work is being redesigned around it.
First-order effects
- More employees and managers now have direct experience with AI at work, expanding the near-term base for productivity-oriented use cases even as many jobs retain existing processes.
- Leaders' more positive assessment can make them the immediate internal advocates for further deployment, while workers who do not use AI regularly remain a significant adoption gap.
Second-order effects
- Software vendors and employers will face greater pressure to make AI useful inside existing workflows, because reported gains have not yet translated into widespread operational redesign.
- The split between leadership sentiment and worker usage raises the importance of implementation, training, and governance; adoption counts alone will be a weak measure of realized workplace impact.
Third-order effects
- If occasional use continues converting into frequent use, workplace AI competition will increasingly turn on daily-use penetration and embedded workflow value rather than simple tool availability.
- The survey points to gradual augmentation rather than abrupt job transformation; whether that becomes structural change depends on organizations converting individual productivity gains into redesigned processes.
The trend: Workplace AI is moving from early, uneven experimentation toward broad but still shallow workflow integration, with sustained value contingent on adoption becoming habitual and operationally embedded.