A Pew Research Center survey suggests many Americans are wary of AI taking a decision-making role in healthcare, especially without human review. In Insurance Journal’s account, 60% of 3,488 US adults surveyed in June said they were uncomfortable with AI helping their doctor diagnose or plan treatment.
Watch Desk analysis
What happened
The survey, as reported by Insurance Journal, found that only 5% of respondents were open to insurance claim decisions being made by AI without human review. Almost half said they did not want AI to play any role in those decisions.
More than half also saw AI as a risk to their personal health information. Respondents said they trusted doctors to use AI in their best interests more than insurers, government, technology firms or hospital leadership.
Why it matters
These results suggest that the question is not simply whether people welcome or reject AI in healthcare. Many may be open to some uses, such as translation, while drawing a firm line at decisions about treatment or insurance without a person checking the result.
That distinction matters as technology companies promote healthcare applications. Transparency about when AI is being used, and who remains responsible for the decision, may be as important to public acceptance as the tool itself.
Our read
The clearest signal here is the demand for human oversight, particularly in insurance decisions. A survey cannot tell us how people will respond to every real-world use, but it does make the “let the AI decide” pitch look like a hard sell. If organisations want trust, they will need to explain where AI fits and where a person still has the final say.
What to watch
- Whether healthcare providers and insurers explain when AI contributes to decisions.
- Whether future surveys find different attitudes towards assistance with review versus decisions made without it.
- How organisations address concerns about personal health information.
Discussion spark: Should AI be allowed to make healthcare or insurance decisions if a human reviews the result, or should people have the right to keep it out of those decisions entirely?
Sources and evidence
- Source update (3 November 2026, 17:08 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.