AI Saves Clinicians Time But Training Gaps Persist, Philips Survey Finds

Artificial intelligence is generating meaningful time savings for healthcare professionals, but the majority of clinicians report that training in the technology remains inadequate, inconsistent or unavailable at their organizations, according to a global survey published Tuesday by Philips (PHIA).

The study, part of Philips' annual Future Health Index, drew on two quantitative surveys — one canvassing 2,011 healthcare professionals and another 20,085 patients — across 10 countries.

Among the most common AI applications reported were using the technology as a sounding board for work-related ideas, transcribing clinical notes and scheduling patient appointments. On the clinical side, use cases included flagging dangerous drug interactions, suggesting diagnoses from symptoms and assisting in the analysis of imaging scans.

Nearly half of surveyed professionals — 46% — reported annual time savings of at least 132 hours on average, while 50% said AI had increased their capacity to see patients. Nurses and doctors said the technology helped them stay current with research and clinical developments and think through cases more thoroughly.

However, organizational adoption has failed to keep pace with the spread of the tools themselves. Some 64% of clinicians said they turned to personal AI tools when workplace options fell short of their needs. Shez Partovi, Philips' chief innovation officer, said the problem was straightforward: "The organizations aren't moving fast enough to provide the tools and the training."

Seven in ten healthcare professionals said AI-related training at their organizations was unavailable, limited or inconsistent. The report called for "structured, role-specific training" to help clinicians build the digital skills and clinical judgment needed to work effectively alongside the technology.

Despite enthusiasm for AI's potential, near-universal caution about its limits was also evident: 90% of professionals said it was essential to keep a human in the loop as AI advances, while 86% said all AI outputs required human oversight.