More older adults are beginning to accept, and support, the use of artificial intelligence in healthcare and data analytics.
The lack of understanding of how AI works, however, continues to fuel suspicions. One ongoing concern consumers have is where AI tools are getting their data from, according to new survey results.
Almost 9 out of 10 the Americans surveyed said they could see problems arising from the lack of transparency around AI data collection, the survey showed. In addition, half of the survey participants said they were concerned about AI producing false information.
The latter is a major issue, because when AI does produce bad results, clinicians and caregivers don’t always notice, and they often double down on the misinformation, a recent study showed.
About half the patients surveyed said that if they knew that their healthcare provider was using AI to diagnose conditions in them, they wouldn’t trust the results.
A slight majority of respondents said they’d accept AI being used in annual screenings, but just 25% said they’d be comfortable with AI being used in surgery.
The survey was conducted recently by healthcare software provider Wolters Kluwer.
“As the healthcare community begins implementing GenAI applications, they must first understand and address the concerns Americans have about it being used in their care,” Greg Samios, Wolters Kluwer president and CEO of clinical effectiveness, said in a statement. “It will take responsible testing as well as understanding the importance of using the most highly vetted content developed by real medical experts to build acceptance of AI in clinical settings.”
At the end of last year, Congress and the Biden administration said that AI transparency in healthcare was something they hoped to tackle soon.
Within senior living, data collection also is tied to residents’ privacy, and providers must make sure appropriate guidelines are in place to protect residents’ rights and avoid liability issues, a healthcare expert noted Thursday for McKnight’s Senior Living.
