After covering the article about how bad AI is at diagnosing ER patients in a recent Doom Hole, I strong-armed Jesse into dedicating a whole episode to the subject. I hope it serves as a warning, and helps you to inform someone in your sphere to ALWAYS seek a real human being when dealing with health issues, mental OR physical.

Let's get into it.

(JH) Today on Substack I experienced a first. I was tagged on an AI-related article about a new law proposed in Utah, or maybe it's just an informal agreement at this stage. The agreement is between an AI company out of San Francisco to provide prescription refills for psychiatric medicines. The need, the state says, comes from overworked doctors, nurse practitioners, and pharmacists, and they see the AI bot to be a relief valve.

I have real concerns about chatbots and human bodies. They mix like oil and water, at least at this moment in time. Chatbots, we are told, have black boxes of data training, programming, weighted algorithms, that we cannot know or see or tweak or question. They have proven to be sycophantic, give us what feedback it thinks we seek, and do not catch subtle cues. The human body, the degree-on-the-wall-nurse practitioner or doctor, is genetically predisposed to catch subtle cues. Thousands of years of fight or flight, of surviving DNA because our ancestors caught a subtle hint of trouble or whatever? It pays off when we interact with other human beings.

In my response to that article, I mentioned Zoloft because it is personal. It can feel like a last-best-lifeline to the patient and when they experience severe side effects, like say, increased suicidal ideation? Scared to have that lifeline yanked away, they may not admit to them. But a human health practitioner? They might see the hesitation and ask a follow up question that could lead to helping the patient find a course of action that is truly lifesaving. A chat bot? Its degree framed on the wall holds a blank piece of paper.

Sources:

Potter Puppet Pals, YouTube

5 Major Disadvantages of AI in Healthcare, Keragon.com

Disadvantages of AI in Healthcare: What CTOs Should Be Aware of, TechMagic.co

Google's healthcare AI made up a body part - what happens when doctors don't notice? The Verge

Man Went to ChatGPT for Health Advice. What He Did Next Led to Poisoning and Psychosis, People Magazine

When Algorithms Fail Medicine: Evidence of AI's Unfulfilled Promises in Healthcare, InfluxMD

Mmm mmm mmm mmm by the Crash Test Dummies - Ger O Donnell & Grainne Hunt Music, YouTube

Send us Fan Mail

Support the show

Support Curious Cat, an independent, human-made podcast!

Anxious about AI? Take two minutes to contact your local politician and ask them to tap the brakes on this technology. Still worried? Contact one of the orgs below and get involved.

But for today, hug your kid, cook food and really breathe in deep as it simmers, walk in nature, brush a cat, donate to the food bank, brew a cup of tea, or draw a five-minute portrait of your dog.

***Is AI the Devil? on Substack!***

Hero Organizations:

80,000 Hours

Center for Humane Technologies

State of Surveillance, an organization that helps foster online privacy

Buy Curious Cat Podcast a Coffee!