Patients are more likely to discuss stigmatising health conditions such as sexually transmitted infections (STIs) with an AI chatbot than a GP, new research suggests.

A study by the University of Westminster and University College London (UCL) found that people preferred to speak to a healthcare professional about severe conditions such as cancer, but were less inclined to do so when it came to less severe conditions and those with a perceived stigma.

The researchers said the study shows there is a place for artificial intelligence-powered chatbots, particularly when the health issue involves the disclosure of personal information that is challenging.

But they warned that the healthcare sector needed also to recognise the limitations of such technology and only use it in certain scenarios.

The research team said chatbots had become increasingly common in recent years within primary care, as artificial intelligence technology had improved, but they argued this had occurred without sufficient evidence for their feasibility and effectiveness.

“Many AI developers need to assess whether their AI-based healthcare tools such as symptoms checkers or risk calculators are acceptable interventions,” Dr Tom Nadarzynski, lead author of the study from the University of Westminster, said.

“Our research finds that patients value the opinion of healthcare professionals, therefore implementation of AI in healthcare may not be suitable in all cases, especially for serious illnesses.”