The attitude towards AI in medicine remains ambivalent, despite its growing use. A recent poll conducted by the analytical firm MAR CONSULT reveals that while many recognize AI’s potential to enhance diagnostic accuracy, most remain hesitant to replace human doctors entirely.
AI in medicine is primarily used in medical image analysis, diagnostics, personalised treatment and telemedicine, as well as in drug development. However, less than half of MAR CONSULT respondents (46%) were aware of its applications in medicine. Awareness was highest among the youngest group (60% of 18-24 year-olds), but dropped sharply to 39% among 25-34 year-olds—the lowest level among all age groups.
More than a third of respondents (39%) view the introduction of AI positively, while almost half (49%) are cautious. Only 7% are strongly negative, preferring entirely human-led diagnostics.
These figures align with a 2023 survey by state pollster VTsIOM (All-Russia Public Opinion Research Center), which found that 40% of Russians were comfortable with doctors using AI for diagnosis and treatment recommendations. Interestingly, the number of people who trust AI has remained unchanged for two years.
Notably, younger respondents are more cautious about AI in medicine than older groups. Among the young, the share of “skeptics” reaches almost 60%. In contrast, enthusiasm rises significantly among those over 45: 56% in the 45-54 group and 45% among those aged 55-65 view AI positively.
Four out of five respondents believe the main advantage of AI is increased diagnostic accuracy and a reduction in doctors’ errors. Nearly two-thirds think AI will save time and speed up medical data processing. Some 42% are confident it can help identify rare pathologies, and 28% believe it will optimise medical staff’s workload.
At the same time, half (51%) doubt AI’s ability to account for individual patient characteristics. Some 46% distrust machine decision-making algorithms, and 36% fear loss of privacy and data leaks. A quarter worry about losing control of the process, and 13% fear a weakening relationship with their doctor.
People are primarily concerned about the depersonalization of the process and the lack of emotional contact with the doctor, says Dmitry Shimanov, founder and chief executive of MAR CONSULT. He added that skepticism also stems from a poor understanding of how AI algorithms work and concerns about diagnostic distortions.
There are areas where respondents are willing to trust AI: 37% would allow it to analyse medical images such as MRI or X-rays, 30% would accept AI-led initial diagnosis, and 18% would trust it to make a final diagnosis. However, only 12% are fully confident in the reliability of AI results, while 70% are doubtful but expect technological improvements. Some 18% completely distrust AI.
When asked about unconditional trust, 22% said they would trust AI provided personal data is protected, and 38% would accept AI involvement if a doctor verifies the results. Trust in a doctor using AI has decreased for 19% of respondents and increased for 12%.
A third of respondents (32%) believe doctors should play the key role, with AI as their assistant. Another 22% say AI is not yet ready to replace doctors but may do so in the future, while a third are convinced AI will never replace doctors.
Some 37% are concerned about doctors being replaced by robotic systems, believing human involvement is key. In the 2023 VTsIOM (All-Russia Public Opinion Research Center) survey, Russians were twice as likely to expect an improvement in doctor-patient relations from AI use (27%) as a deterioration (13%).
Methodology: The research was conducted online with 600 participants across Russia in July 2025.
