Dr. Mehmet Oz

We actually understand if a human being who has an MD degree makes a mistake, we think AI is perfect.

and are completely intolerant.

They've been in a minor transgression.

What you said earlier is true, and these studies have been done several times now.

The large language models do better on board exams.

So general knowledge, they're better than a doctor.

They're more patient than a doctor.

If you talk to an AI-informed avatar, they'll answer the same 10 questions on diabetes all day long and not get bored.

talk to a general practitioner after they've talked to 10 patients with diabetes who've asked the same 10 questions you get a little terse in your responses that's just human nature i don't want to answer the same question every time but that's what patients need you to do so i think we have the opportunity and we looked at some of these models in fact john doerr brought a bunch of very thoughtful entrepreneurs into the building to talk about the possibility of us taking the average general practitioner and making them five times maybe even ten times more efficient

Now, here's the big deal.

There are not enough GPs in America.

We probably need two to three times more just to keep up with Europe.

It's just per capita.

We're way below because my son just finished medical school.

There aren't many kids in medical school at Columbia where he went who are going to become GPs.

They all want to be ophthalmologists and orthopedic surgeons because they pay a lot more.

And this is even, Columbia is free.

If you don't have money, you don't have to pay tuition at Columbia.

They did that so that kids would go into general practice and hasn't had the desired impact yet.

But the broader question is, how do we allow AI to play a role as this transition period happens so people feel confident?