Adults in the United States who regularly use artificial intelligence chatbots for health advice are more likely to believe false claims about vaccines, according to a new survey released on Tuesday by health research firm KFF.

The poll, conducted in May among a nationally representative sample of 2,480 U.S. adults, found that frequent AI users were more likely to believe misinformation, including the debunked claims that vaccines cause autism or that the measles vaccine is more dangerous than the disease itself.

The link remained even after accounting for factors such as age, race, education and political affiliation.

Researchers and public health officials have long warned that AI could amplify misinformation and shape public opinion. The findings come as more Americans increasingly rely on AI chatbots for medical guidance.

A separate KFF survey conducted in March found that about one in three U.S. adults had sought health advice from AI tools.

AI companies have also acknowledged that a significant number of users turn to their chatbots for medical information.

“Health is already one of the most common ways people use ChatGPT, with hundreds of millions of people asking health and wellness questions each week,” OpenAI said in a January blog post.

According to KFF’s survey, 35 per cent of U.S. adults who use AI tools for health advice at least once a week said they believe it is “definitely or probably true” that the measles, mumps and rubella vaccine has been proven to cause autism in children. By comparison, 29 per cent of those who use AI for health information only occasionally held the same belief, while the figure dropped to 20 per cent among adults who do not use AI for health advice.

The debunked claim linking the MMR vaccine to autism remains a central narrative of the anti-vaccine movement, which has gained renewed prominence following the COVID-19 pandemic and the appointment of Robert F. Kennedy Jr. as U.S. Health Secretary.

The false claim linking the MMR vaccine to autism gained widespread attention after a study published in The Lancet in the 1990s.

The paper was later fully retracted after its findings were found to be fraudulent, and numerous subsequent studies have found no evidence of a link between the MMR vaccine and autism.

KFF’s survey also found that 29 per cent of U.S. adults who frequently use AI tools for health information believe the false claim that mRNA vaccines can alter a person’s DNA, compared with 20 per cent of those who never use AI for health advice.

In addition, 22 per cent of frequent AI users said they believe the measles vaccine is more dangerous than the virus itself, versus 15 per cent of non-users.

The survey also found a similar pattern among people who rely on social media for health information. According to KFF, adults who use social media for health advice at least once a week were more than twice as likely as those who do not to believe the debunked claim that the MMR vaccine causes autism, with 37 per cent saying the myth is “probably” or “definitely” true, compared with 16 per cent of non-users.