A Stanford study reveals a significant gap between what American workers want from AI and what the technology can deliver.

Researchers from the Stanford Institute for Human-Centered AI and the Digital Economy Lab surveyed 1,500 workers across 104 occupations and interviewed 52 AI experts to assess worker expectations, technical capabilities, and potential workplace applications. Their findings suggest that most workers welcome AI support for tedious or repetitive tasks. However, they remain skeptical of its use in more complex or creative responsibilities.

Trust emerged as a major concern: nearly half of those surveyed questioned AI’s accuracy, while 23% feared job loss and 16% worried about the absence of human oversight. Workers were especially hesitant to delegate creative work or sensitive communications to AI. Still, most welcomed automation in administrative tasks such as scheduling appointments, correcting records, and organizing information. They believe those areas are where AI can enhance efficiency without replacing human judgment.

The study found that 45% of workers preferred a collaborative partnership with AI. Meanwhile, 35% favored human oversight at key decision points. Fully automated systems were the least desirable option.

“This study gives us a roadmap to align future AI systems with what people actually want,” said co-author Erik Brynjolfsson, director of the Stanford Digital Economy Lab. “AI should support people, not sideline them.”

Using input from AI experts, the researchers sorted workplace tasks into four categories: those well-suited for AI, those workers don’t want automated, those with high demand but low technical feasibility, and low-priority tasks. Notably, 41% of current AI deployments fall into categories that workers find unhelpful or untrustworthy — suggesting that many companies may be misaligning their automation strategies.

The study also points to a shift in the skills most valued in the AI age. Analytical and data-driven roles may lose prominence, while jobs requiring communication, teaching, and human empathy are likely to grow in importance.

Lead author Yijia Shao emphasized that including workers in AI design decisions is essential for trust and long-term success. “This isn’t just about efficiency — it’s about building systems people want to work with.”

Need Help?

If you’re concerned or have questions about how to navigate the AI regulatory landscape, reach out to BABL AI. Their Audit Experts can offer valuable insight and ensure you’re informed and compliant.