Why Most Medical AI Is Garbage—And Why No One Cares
🚀 The historic launch of my new podcast, Digital Health Inside Out.
Welcome to AI Health Uncut, a brutally honest newsletter on AI, innovation, and the state of the healthcare market. If you’d like to sign up to receive issues over email, you can do so here.
🎉 I’m beyond excited! 🚀 The groundbreaking inaugural episode of my new podcast, Digital Health Inside Out, with my friend and co-host, Alex Koshykov, was recorded on January 13, 2025—and it’s now live for everyone!
The theme of this very first episode was “Why Most Medical AI Is Garbage—And Why No One Cares.”
Please watch, support our efforts, and join the conversation. Discussions drive the truth forward.
Four speakers for this episode:
⭐ Alex Koshykov, CEO @ YODD and COO @ BeKey
⭐ Dr. Hugh Harvey, Managing Director @ Hardian Health
⭐ Sergei Polevikov, Founder @ WellAI
⭐ Tiya LaCroix, Sr. VP of Healthcare Innovation & Growth @ Impensa
Here are the most headline-worthy quotes from this hour-long podcast. Enjoy!
On the State of AI in Healthcare
“Because let’s face it, I think one of the most important pieces that is missing in healthcare is transparency, and that’s what we’re going to try to change.” —Alex Koshykov
“AI is the next shiny object that we think is going to fix a lot of our issues.” —Tiya LaCroix
“Anytime somebody says, ‘Oh, well, let’s just slap some AI on top of this,’ and I’m like, well, we can’t even fix the underlying data problems to be able to match a PO to an invoice.” —Tiya LaCroix
“Healthcare is a very high-risk environment. Everyone around the world relies on healthcare, and they trust, hopefully, in the physicians and the nurses and the people providing that care.” —Dr. Hugh Harvey
“AI models weren’t designed to be logic reasoners or fonts of knowledge. They were designed to transform text.” —Dr. Hugh Harvey
On the Risks and Limitations of AI in Medicine
“Hallucination is a big problem—unfortunately, that’s one of the drawbacks of the attention mechanism in transformer models.” —Sergei Polevikov
“The general data training set contains a lot of what we would consider, I guess, poisonous to a large language model.” —Dr. Hugh Harvey
“In healthcare, we definitely don’t have a single source of truth.” —Tiya LaCroix
“Medicine is not an exact science. It’s largely experience- and judgment-based.” —Dr. Hugh Harvey
“Even a percentage point difference in a wrong result can be a matter of life or death for a patient in the end.” —Dr. Hugh Harvey
“A doctor knows when they don’t know something. AI doesn’t.” —Dr. Hugh Harvey
“If I can’t tell you how much those drugs actually cost to deliver, how do you expect AI to automatically be able to diagnose and recommend a treatment for cancer?” —Tiya LaCroix
“Patients don’t present like this. They don’t present with a nice, neat list of well-defined phenotypes written in a medical language.” —Dr. Hugh Harvey
On Why AI in Healthcare Keeps Failing
“Compute always wins.” —Sergei Polevikov
“If we imagine a future where the tech bros win and the whole of medicine is going to be passed over to the logic of LLMs, you’re going to have LLMs talking to LLMs talking to LLMs the whole way down.” —Dr. Hugh Harvey
“If they are all hallucinating and talking to each other and responding and replying, I don’t think it will take many generations of that to cause absolute havoc with the sanctity of healthcare.” —Dr. Hugh Harvey
“You still can’t send scans from Scotland down to London that simply.” —Dr. Hugh Harvey
“I can see some administrator pulling you guys into a meeting where you’re having to defend your diagnosis versus AI.” —Tiya LaCroix
“The migration of my grandfather’s healthcare data did not happen through the healthcare systems to the final system.” —Tiya LaCroix
On AI Business Hype (Hippocratic AI, AI Scribes, VC Grift, and Broken Incentives)
“A $9 an hour AI nurse? That’s not innovation—that’s an insult.” —Tiya LaCroix
“We all know [the Hippocratic AI latest funding rounds are] bullshit.” —Sergei Polevikov
“We all know VC investing a big pump and dump scheme.” —Sergei Polevikov
“Hippocratic AI is the glorified example of the broken system.” —Sergei Polevikov
“If you hear about a “health AI revolution,” check who’s making money off it first.” —Sergei Polevikov
“AI scribes are helpful, but many are copycats and AI tourists.” —Sergei Polevikov
“So it just happened that Hemant at General Catalyst—he’s considered sort of, you know, an expert in healthcare, even though every time he touches healthcare, it seems like he creates more destruction.” —Sergei Polevikov
“Munjal started Hippocratic AI with his buddy Hemant and Andreessen Horowitz, who, you know, just blindly give money to all of their buddies.” —Sergei Polevikov
“You can become a billionaire by screwing everybody else.” —Sergei Polevikov
“As soon as the nursing organizations start hearing more and more about these $9 an hour AI nurses, they’re going to organize and put a kibosh on this.” —Tiya LaCroix
On What Needs to Change
“We don’t need AI in healthcare—we need automation that actually works.” —Sergei Polevikov
“For most of the medical necessity and tests from what I’m observing daily with my medical colleagues—you don’t need AI. You just need good ol’ automation.” —Sergei Polevikov
“We’re talking about AI like we’re in the 1990s here in the U.S.” —Sergei Polevikov
“Healthcare is 20 years behind in terms of some of these technologies.” —Sergei Polevikov
“If you’re a patient in a rural area and you’re 100 miles away from the nearest medical facility, talking to AI could be useful as opposed to not having anything.” —Sergei Polevikov
“Tech in healthcare should be about making nurses’ and doctors’ lives easier, not completely removing them from the equation.” —Tiya LaCroix
“I don’t want to remove the actual care from healthcare.” —Tiya LaCroix
You can access the full 1h 16m episode here.
For those eagerly awaiting my mind-blowing investigation into Hippocratic AI—thank you for your patience. I promise it’ll be worth the wait. Stay tuned.
👉👉👉👉👉 Hi! My name is Sergei Polevikov. I’m an AI researcher and a healthcare AI startup founder. In my newsletter ‘AI Health Uncut,’ I combine my knowledge of AI models with my unique skills in analyzing the financial health of digital health companies. Why “Uncut”? Because I never sugarcoat or filter the hard truth. I don’t play games, I don’t work for anyone, and therefore, with your support, I produce the most original, the most unbiased, the most unapologetic research in AI, innovation, and healthcare. Thank you for your support of my work. You’re part of a vibrant community of healthcare AI enthusiasts! Your engagement matters. 🙏🙏🙏🙏🙏