The Trump administration is paving the way for AI chatbots to provide diagnoses and medication to patients, but experts have already warned the technology could be fraught with problems.
Amy Gleason, the head of DOGE and supporter of the Make America Healthy Again (MAHA) movement, became a proponent of AI diagnoses after her daughter Morgan received a life-changing diagnosis from ChatGPT.
Morgan, 27, had spent years battling a debilitating autoimmune disorder but when she uploaded 16 years of medical records the AI chatbot told her she had a different condition to the one doctors had diagnosed.
This ultimately allowed her access to a clinical trial, and won her mother’s support for the benefits of artificial intelligence in health.
“People are seeing the difference the AI is bringing,” Gleason said. “And it’s like the genie is out of the bottle.”
The Trump administration is now looking at a number of ways to integrate such technology into America’s healthcare system, including offering $50 million in research awards to develop AI software that can deliver cardiovascular care, The Washington Post reports.
That could mean if someone calls a medical provider with symptoms of a heart attack, they are dealt with by a chatbot.
The administration is also backing a three month trial program in Utah, which would allow AI to refill prescriptions for patients.
According to The Post there are ongoing internal conversations in the White House looking into regulating independent AI doctors, comparing them to self-driving cars.
Gleason championed the benefits of AI technology in an address to GovCIO Media & Research’s Federal IT Efficiency Summit in May.
Referring to her daughter’s case, Gleason said: “She didn’t have to know the right question to ask, and she didn’t have to know all of her information or what was in her biopsy, it just did that for her.
“That’s the power of having that data and having technology with it.”
Other backers, including the administrator of the Centers for Medicare and Medicaid Services Mehmet Oz, believe the move towards AI will help alleviate the lack of rural doctors.
There are hurdles to overcome, however. Currently, AI is only legally able to offer medical guidance with legal disclaimers attached. It cannot yet practice medicine, per laws in place by the Food and Drug Administration (FDA) and state boards.
Critics are concerned about AI’s ability to work unchecked, citing an Oxford Internet Institute study from February which found that OpenAI and Meta’s AIs GPT and Llama were “no better” at providing diagnoses than human doctors.
“Despite all the hype,” the study states, “AI just isn’t ready to take on the role of the physician. Patients need to be aware that asking a large language model about their symptoms can be dangerous, giving wrong diagnoses and failing to recognise when urgent help is needed.”
And others are sceptical over the ties between the White House and Silicon Valley, and question whether AI is really smart enough to work in such a sensitive area without human oversight.
Robert Wachter, author of “A Giant Leap,” a book on AI and medicine, told The Post, “You’re combining a general anti-regulatory, pro-business administration with very close ties to an enormous amount of wealth to a segment of society that wants us to go fast.”
The White House and the FDA have been contacted for comment.
Healthcare is not the White House’s only push for AI, as the president recently signed an executive order to create a framework for vetting artificial intelligence.
This executive order is aimed at determining AI national security risks. It would give the government 30 days to review a new AI system for safety. This short time frame is, per the White House, in place to keep the U.S. ahead of other nations in the AI race.
Leading AI companies OpenAI, Anthropic, and Google have all described this policy as an important step in the U.S’ use of the technology.
More details here...
