Dr. Google may have some competition
I came across a recent Pew Research Center report looking at how Americans are using AI chatbots for health. It seems like we're asking GenAI platforms such as ChatGPT about a lot more than what to make for dinner and what does it mean when my dog looks back at me when we go on walks?
Honestly? I'm not surprised.
Seeking answers on the internet is not a new concept. For better or worse, we've been Googling our symptoms. First, we asked Jeeves (again, another unique and niche millennial references. IYKYK). Then we googled and WebMD'd (not a verb, but you get it.) In fact almost 73% of Americans reported turning to the internet FIRST for health or medical information (HINTS, 2019). That's substantially more than books. More than family and friends. More than ACTUAL health care providers!
I guess the difference now is is that instead of sorting through 17 search result pages, we can now ask the question, get an answer(potentially fabricated), ask a follow-up(again most likely not completely accurate due to hallucination) say "can you explain this to me like I'm five?"
There's something potentially really valuable about that from a health literacy perspective, but where I worry is when it goes past understanding symptoms, diagnoses and lab results and extends to folks asking for help deciding whether something warrants a trip to the doctor. We've known for a long time that health information isn't always written for the people who actually need to understand it, but accessibility doesn't necessarily equal accuracy.
That is the interesting layer. It's no longer just "can you find, understand and use health information?" We have to also think about whether people know how to evaluate information generated specifically in response to their questions and recognize when AI isn't the appropriate source.
