Dive Brief:
- More than 40 million people ask ChatGPT healthcare questions every day, signaling consumers are frequently turning to the chatbot to navigate the complex healthcare system, according to a report published this week by OpenAI.
- Roughly 1.5 million to 2 million questions are asked about health insurance each week, including on how to compare health plans, handle claims and billing, and understand prices, said OpenAI, the maker of the popular chatbot.
- About 7 in 10 health-related conversations with ChatGPT take place outside typical clinical hours, suggesting users are looking for information when they can’t readily access their providers, the report found.
Dive Insight:
Artificial intelligence is an enticing emerging technology for the healthcare sector, igniting hopes that the tools can take on the industry’s hefty burden of administrative tasks and alleviate workforce shortages.
As healthcare professionals become more frequent AI users, their patients are turning to AI chatbots too. More than 5% of all ChatGPT messages across the globe are about healthcare, averaging billions of messages each week, according to OpenAI.
Healthcare is a pain point for many Americans, given the country’s complex and often confusing insurance landscape and access challenges. According to an OpenAI survey taken in December, 3 in 5 U.S. adults say they’ve used AI tools for healthcare or health questions in the past three months.
Fifty-five percent said they used AI to check or explore symptoms, while 48% reported they used the technology to understand medical terms or instructions. More than 40% said they used AI to learn about treatment options.
Users are also often turning to AI to ask health questions when providers aren’t available. That’s a more prevalent challenge for people in rural communities, who have worse health outcomes compared with their urban counterparts and face longer travel times to providers.
Late last year, ChatGPT received an average of more than 580,000 healthcare-related messages per week in hospital deserts, or areas that are more than a 30-minute drive from a general medical or children’s hospital, according to the OpenAI report.
Still, patients have to contend with the potential risks of the technology, including erroneous responses. AI chatbots can sometimes hallucinate, or give incorrect or misleading answers to questions that patients might not easily recognize.
Meanwhile, the Trump administration has largely adopted a deregulatory stance toward AI. States have moved to fill the gap, introducing legislation on the technology in all 50 states last year, according to the National Conference of State Legislatures.
But President Donald Trump in December also took aim at states’ growing push to regulate AI, signing an executive order that challenges some state laws and calls for a national framework that would preempt state actions.
