(WXYZ) — A lot of us have done it; you experience a symptom, maybe something new or concerning, but instead of calling your doctor, you ask Artificial Intelligence. It's fast, it's easy, and for many people, it’s becoming a go-to source for health advice. But Americans are becoming increasingly wary of AI's role in medicine.
Watch Keenan Smith's report in the video below
Artificial intelligence is becoming part of everyday health care, but a new national survey shows Americans may be becoming more skeptical and less trusting of AI. While others are skipping checking in with their doctor to self-diagnose with the help of that very same artificial intelligence.
From lab results to late-night symptom searches, AI is becoming a go-to tool for patients looking for quick answers. But new data from The Ohio State University Wexner Medical Center shows trust in that technology is slipping. Despite initial excitement about the problem-solving power of AI, support for using artificial intelligence in health care has dropped to just 42 percent, down from 52 percent two years ago.
Dr. Hiten Patel is a family physician at The Ohio State University Wexner Medical Center. He says that early excitement is now giving way to genuine concerns.
"As it's kind of pervasively been in all parts of our life and our personal life and our health and in our work lives, we've seen that it does have limitations," Dr. Patel said. "There are safety and privacy concerns. So about 70 percent of Americans felt that they were concerned about privacy with AI in their use and AI use in their healthcare."
The concerns include what happens with their personal information shared with AI, how long that information will be held by the AI model, and how will it be used.
There may also be a growing realization about what AI can and cannot do. And the reliability of those answers, no matter which AI model a patient uses. All current AI models are subject to hallucinations - making up facts and taking unfounded logical leaps that can make AI answers unreliable.
"And so I think a lot of people are seeing this. Um, when they use AI for healthcare, but they're also seeing some risks in terms of security. So they're becoming more hesitant in using AI," Dr. Patel said.
Even with those concerns, other Americans are still using AI as part of their health decisions, often before ever speaking with a doctor. For some, it is a preference; for others, it fills a gap in an already strained healthcare system.
"Challenges getting into your doctor, there are challenges getting answers from doctors because of how busy the healthcare system is and how complicated it is," Dr. Patel said.
How are people using AI for health care? According to the survey:
- 62 percent use AI to help understand symptoms before deciding whether to seek medical care
- 44 percent use AI to help explain test results or a medical diagnosis
- 25 percent use AI to compare treatment options or help make a treatment decision
- 20 percent use AI to prepare for an upcoming medical appointment
While AI can be helpful in explaining terms or in some cases interpreting test results, Dr Patel says there’s still a critical difference between AI and a human provider.
"We have a lot of context of a patient, kind of getting back to preferences, like what they prefer, how likely they are to be amenable to a certain treatment or open to a treatment, and what their past medical history is," Dr. Patel said.
Instead of replacing doctors, experts say AI may be most useful as a tool to help patients prepare before and after care, by helping patients recall symptoms, remember when those symptoms started or how to put your lifestyle recommendation into practice, like exploring weight-loss or diet options.
That’s the safest approach is to use AI; as a starting point or as a companion, but not the final answer. And remember to always loop in a medical professional when it comes to your care.
