I really dislike this guy and even more question his ideas.Arificial intelligence is not ready for prime time in medicine. Here is the article:

Dr Mehmet Oz reportedly told federal staffers that artificial intelligence models may be better than frontline human physicians in his first all-staff meeting this week.

Oz told staffers that if a patient went to the doctor for a diabetes diagnosis it would cost roughly $100 an hour, compared with $2 an hour for an AI visit, according to unnamed sources who spoke to Wired magazine. He added that patients may prefer an AI avatar.

Oz also spent a portion of his first meeting with employees arguing they had a “patriotic duty” to remain healthy, with the goal of decreasing costs to the health insurance system. He made a similar argument at his confirmation hearing.

“I think it is our patriotic duty to be healthy,” Oz said in response to a question from the Republican senator Todd Young of Indiana. “First of all, it feels a heck of a lot better … But it also costs a lot of money to take care of sick people who are sick because of lifestyle choices.”

a woman with a sign saying 'I'm for medicaid'
Medicaid recipients demand that Congress protect healthcare program
Read more
Oz was recently confirmed by the Senate to lead the Centers for Medicare and Medicaid (CMS) – one of the largest and thorniest of the federal government agencies. CMS oversees $1.5tn in annual spending, health insurance for nearly half of Americans and incredibly dense regulations that underpin much of America’s health insurance system.

His confirmation came as the federal Department of Health and Human Services is slated to lose roughly a quarter of its workforce amid cuts by the HHS secretary, Robert F Kennedy Jr, and the billionaire Elon Musk’s unofficial “department of government efficiency” (Doge).

CMS has seen a relatively small portion of the cuts – 300 workers compared with a total loss of 20,000. However, Republicans are proposing huge cuts to Medicaid, the health insurance program for the low-income overseen by CMS, which insures roughly 71 million people. Republicans are seeking to cut the program to pay for tax cuts that overwhelmingly benefit the wealthy.

The 64-year-old was a surgeon before hosting The Dr Oz Show, a spinoff of Oprah Winfrey’s talkshow. He hosted the program for more than a decade, and ended it to seek an open Senate seat in Pennsylvania. He lost to the Democrat John Fetterman in 2022.



Oz has long been criticized for his endorsement of unproven cures – among them “magic” weight loss supplements and his endorsement of hydroxychloroquine to treat Covid-19.

link https://www.theguardian.com/us...mehmet-oz-doctors-ai

In the distant future, perhaps. Now, no. We can test it on him though.


By the way, who gets sued for malpractice when AI is wrong?

In a way, this could be done. Some of it is an FDA issue, some of it is legal, some of it is technical.

Modern medicine can be very close to witch doctoring, at times.

In theory, a rapid testing set up, and an automated kiosk to process blood/urine/stool/hair/maybe saliva could really help.

Near future or distant, like it or not, it will happen. I'm just glad that I will be long done by then.

Caveat I’m working on a project in this area:

The goal of this, is to get everyone a high quality diagnosis. (At least the goal of the team I’m on an advisory board for.)

Every doc we have talked to, seems quite happy about it. There’s so much information to process and keep up with

I can definitely see the usefulness in having AI share information across multiple platforms. This would allow doctors to quickly analyze data from all sources and to cross reference results findings and diagnosis.

But taking out the human rationalization of seeing a patient as a whole rather than a list of issues is no good. Sadly, this is currently the way a lot of doctors see patients. They are too quick. You get a 10 minute visit and a bottle of pills.

So maybe AI could alleviate some of those issues doctors have if used as a tool and not a crutch. AI could do a lot of the sorting and offer conclusions but the doctor would then have to time to look more at the person as a whole and finalize the experience.

He is not wrong.
You seem to be presuming he is talking about tomorrow, which he is not.
I am 42 it will certainly be within my lifetime at what point can’t tell yet.

This will go for many non litigating attorneys as well.

quote:

Originally posted by konata88:
In the distant future, perhaps. Now, no. We can test it on him though.


By the way, who gets sued for malpractice when AI is wrong?

There will still be a provider and or practice behind the AI and they will have insurance. So they will likely be the 1st to be sued by the patient or their family.

The provider, practice, or their insurance company will then sue the AI software company.

I think in the absence of any care, this is true. In the third world, it will be a godsend. AI supported non-physician clinicians can do a lot. Fairly soon.

In the U.S., I think AI will also support, but it will be slowed, if only by regulation.

As for litigation, I imagine the lawyers will build well-buttoned up EULAs that people will click without reading.

1. If this is an emergency dial 911
2. This information is best effort. Consult with your physician for an authoritative diagnosis and treatment plan.
3. By using this service, you release Dr. AI and it's subsidiaries from all liability.

✓ I AGREE

And even if it were that concise, most people wouldn't even read that.

For an "annual check up" or most routine care, I'll bet it happens fairly quickly. Harvard Med did a study a few years ago showing that an annual physical (in the absence of symptoms) had no effect on all cause mortality. Considering the current format of such visits, I believe it. A few minutes to look at lab results that have already been graphed and flagged, a few questions and 1 minute listening to heart and lungs is extremely unlikely to uncover anything that wouldn't have been caught by AI going through those same questions and numbers.

quote:

Dr Mehmet Oz reportedly told federal staffers that artificial intelligence models may be better than frontline human physicians in his first all-staff meeting this week.

Is there a chance that AI won't just sing the company tune?

Ever since Obamacare, the hospitals bought out all of the independent doctors. During Covid, I had my eyes opened to it. The doctors no longer think for themselves. They are pre-programmed to go along with whatever PHarma the employer wants pushed. I no longer trust the medical establishment.

quote:

Originally posted by chellim1:

Ever since Obamacare, the hospitals bought out all of the independent doctors.

"All the independent doctors?"

If I've told you once, I've told you forty million times: don't exaggerate.

My primary care doc has an independent practice, as do the last two urologists I have seen, as well as the ophthalmologists, dermatologist, and physiatrist (not a typo). Matter of fact, most of the physicians that I am familiar with in this area, are independent.

quote:

1. If this is an emergency dial 911
2. This information is best effort. Consult with your physician for an authoritative diagnosis and treatment plan.
3. By using this service, you release Dr. AI and it's subsidiaries from all liability.

4. "Do you agree to pay for ANY charges valid or not of ANY amount not covered by your insurance" if not, the doctor cannot see you.

quote:

"All the independent doctors?"

OK, sorry. Maybe not all...

Maybe in the 24½th century we'll have holographic doctors. But for now, there are some things AI shouldn't be doing. Alluded to above, what's your recourse if it screws up?

DEI has infiltrated medicine too. I dont know if you guys realize it, but medical schools have been lowering test standards and rejecting high scoring individuals in favor of anything but.

And this has already been going on for years and years.

When is the last time you saw a GOOD doctor who wasnt a dinosaur? One who graduated this century?

I have NOT. Gimme Grok now!

How we'll get the testing for the AI in the future

I work with AI implementation daily and completely agree that in the current LLM form its GREATLY overhyped. However, It does some things very well and with the right model it could easily replace average GP visit with a basic blood draw based and AI evaluation of the results (based on my less than stellar experiences in the Mayo system)

Currently Appointments are months out - test results are posted to my chart in hours. I look the results up online and can easily find out what it means and what the common treatments are, including side effects. For instance - my blood pressure has consistently been border line ever since I was a child. It was fine until I turned 45 - then it had to be treated - check that box since insurance will now pay for it. AI can easily evaluate basic if-then statements.

Physical exams - we're not there yet. Imaging? damn near there. Blood test interpretation? absolutely. Acute mental health episodes (daughter had some serious issues during covid)? Well anything has to be better than Mayo Rochester and their 'teaching hospital' bullshit.

quote:

Originally posted by konata88:
By the way, who gets sued for malpractice when AI is wrong?

Same as what will happen when 'self-driving' cars are involved with injury/death/property damage.

Ask the trial lawyers and they will say, 'anyone and anything remotely connected that has deep pockets or decent insurance.'

So, to answer the question: the AI bot programmers, whoever hosted the AI bot, anyone whose data was provided as training to the AI bot, any medical professionals who might have referred anyone to the AI bot, and all organizations involved with the above.

And, of course: