How do nurses feel about AI?

For the report, Elsevier surveyed 2,757 clinicians, including 692 nurses and 2,065 physicians, from 118 countries about their work and AI usage between December 2025 and February 2026.

When it comes to their work, 61% of nurses said they are seeing more patients than the year before, and 71% say they have enough time to provide good care — higher than the 60% of physicians who said the same. However, nurses who said they didn't have enough time highlighted high patient volumes (71%), excessive administration (64%), lack of support staff (56%), and increasing complexity of patients' medical needs (56%) as the primary challenges.

Nurses also reported feelings of burnout. Around a third of nurses said they experienced tiredness or exhaustion that impaired their ability to treat patients effectively. Over 40% of nurses also said keeping up with medical advances or new technologies is challenging.

Many nurses are supportive of AI, but their usage and trust of the technology is still lacking. Eighty percent of nurses said AI wouldn't replace clinicians but would become a critical assistant in the next five to 10 years. Similarly, 78% of nurses said AI skills will be essential to clinician training, and 61% expect clinicians using AI to provide better care than those who don't use the technology.

However, only 41% of nurses report using AI at work — much lower than the 57% of physicians who reported the same. Among nurses who use AI, most use generalized AI tools for professional education (42%), patient education (40%), and querying medical literation (34%). Only 30% of nurses frequently use a clinical-specific AI tool, compared to 37% of physicians.

Many nurses also say that they are often not included in decision-making about technology at their organization. Over 40% of nurses said their views are rarely or never adequately represented in their organization's decision-making processes. In comparison, only 19% of physicians said their views are rarely or never represented.

There also continues to be significant distrust around AI tools. Only 42% of nurses say that AI tools are trustworthy. Nurses are also less likely than physicians to believe that AI will save them time in the next two to three years, or that AI will routinely analyze medical images.

To increase trust in AI solutions, nurses said that the tools need to be easy to use (65%), comprehensive across multiple sources (62%), transparent with citations (61%), safe and unlikely to cause harm (60%), and created using peer-reviewed data (60%). 

Amy Hall, dean of the School of Nursing at Franciscan Missionaries of Our Lady University, also highlighted the need to include nurses in decisions around AI and its implementation.

"Nurses need a stronger voice in which tools are adopted to support patients and how they are implemented," Hall said. "When nurses aren't included in AI decisions, it's harder to integrate these technologies into everyday practice in ways that truly support patients and workflows."

"If technology is introduced that positively impacts nurses' workflow and patients' outcomes, nurses are likely to adopt it," said Monica Westhead, a managing director at Advisory Board. "However current AI applications don't fit seamlessly into a nurse's responsibilities and can require significant changes in the way nurses practice."

For example, Westhead noted that nurses are not taught in school to dictate all care being delivered, something that is essential for AI-based ambient listening documentation. Most AI tools are also developed for physicians, and some may be adapted from physician workflows, which are not always compatible with nursing workflows.

"Nursing workflows — and the experience of working as a nurse — differs significantly from physicians," Westhead said. "It is incumbent on AI market leaders to co-develop solutions with nurses and nurse leaders that serve the goals of healthcare provider organizations, specifically supporting nurses."

Similarly, Ali Knight, an RN and research consultant at Advisory Board, said that "[a] successful tech implementation — one that solves the identified challenge — needs to work for the care team and no one understands nurses' needs better than nurses."

"The product itself needs to be intuitive and make patient care more reliable, easier, or faster," Knight added.

"At a time when healthcare systems are under increasing pressure, clinicians see clear potential for AI to improve patient care and efficiency," said Jan Herzhoff, president of Elsevier Health. "However, realizing that potential requires more than access — it demands trusted, evidence-based tools, proper training, and inclusive implementation that supports all members of the care team."