Ezra Klein interviewed NVIDIA CEO Jensen Huang on his podcast last week. At the very beginning of the discussion, Huang’s example of what he describes as the “applications layer” of AI is radiology:
Huang: In the last 10 years, since computer vision really became, if you will, superhuman, A.I. technology has now permeated all of radiology.
Every single radiology application has A.I. in it. As a result, you can detect any anomaly. You can detect any disease, and it does it at a superhuman level.
Klein: Radiology is an example I know you like to use. The thing people worry about with the applications layer is that these applications are going to replace human beings.
Radiology has been an interesting example used on both sides, and I hear you talk of it often. So how has the entrance of A.I.-aided radiology shifted radiology as a practice?
Huang: The thing that’s important to recognize is that for everybody’s job, there’s the purpose of the job and then there’s the task you do as the job.
In the case of radiology, the task — and it consumes a lot of their time, and they sit in dark rooms doing it a lot — is to study these scans.
Now, if all of a sudden the studying of the scan is done automatically, it doesn’t change the purpose of their job, which is to diagnose disease, help doctors do more scans, ultimately help patients figure out what’s wrong with them.
So the fundamental purpose doesn’t change. The task of studying that scan has become automated.
As a result, radiologists are actually able to do more, handle more cases, do more scans. Hospitals are able to process a lot more of these patients, and therefore their revenues go up. As a result, they need more radiologists. So this flywheel is happening because the pipeline of patients is quite large.
The task vs. job formulation has gotten rightfully popular in AI discussions over the past year, but otherwise, Jensen Huang continues to live in a fantasy.
I wrote about Huang being wrong about this last year in November (Hallucinating about AI) and December (Radiology Isn’t an Example of Jevons Paradox), and then another one when Anthropic’s Dario Amodei did the same thing this past March (Dario Dreams of Electric Radiology).
Saying the industry has “largely been converted to AI-driven radiology” is, at this point, simply a lie. I do not believe the man who runs the largest company on the planet (market cap of $5.5 trillion!) and lynchpin for basically everything AI-related has not learned the truth. It is, ultimately, just a compelling, comforting, convenient narrative that sounds plausible. Alas, the past decade has not been, as he and Dario have tried to suggest, a soothing example of Jevons paradox. (For those just joining, those three posts serve as a helpful explanation for how comically their talking points misrepresent our shared reality).
I wish Klein and other journalists would push back more. It is a missed opportunity and cheapens/undermines the exchange. A deep dive podcast is a great venue for giving in to tired soundbites and specious arguments.
Klein and Huang return to the radiology topic again toward the end of the interview to discuss predictions, referring, of course, to deep learning pioneer Geoffrey Hinton’s famous 2016 doozy:
I think if you work as a radiologist, you’re like the coyote that’s already over the edge of the cliff but hasn’t yet looked down so doesn’t realize there’s no ground underneath him. People should stop training radiologists now. It’s just completely obvious that within 5 years, deep learning is going to do better than radiologists because it’s going to be able to get a lot more experience. It might be 10 years. But we’ve got plenty of radiologists already.
Of course, Hinton was not just wrong ten years ago but also unhelpfully so. For the record, here is what I thought in 2023 (feel free to check in 2033 to see how my commentary holds up). Klein again takes the historical argument—which is totally false—at face value, but if you ignore that part, Huang makes a strong but frankly self-incriminating point:
Klein: Well, let me take the side of this to give voice to the fears people have.
There is the example of the radiologist, which people were, over the past 10 years, predicting that job would go away. And right now, there’s more demand for it than ever.
There’s also the reality that automation does wipe out jobs. If you look at today versus 1960, fewer Americans work directly in manufacturing than did in 1960, and we are a much bigger country.
Huang: Let’s take it at face value that the recommendation is exactly what he said, which is that nobody should want to be a radiologist and the world has no radiologists today.
Is that helpful or hurtful to society?
I think we can all agree — we can both agree it would be terribly hurtful. It didn’t happen.
Is it good or bad that we scare young people about the future of A.I., so much so that they don’t even want to go to universities and don’t want to go to college anymore because they don’t think they’ll get a job? Is that helpful or hurtful, if it were to happen? It’s hurtful.
Don’t think for a second just because you’re an alarmist that you’re doing a social good. It is not true.
So I think that we ought to just all be wiser, more mature, be evidence based, be scientific. If you want to be scientific, be scientific. Do the science. But alarming people, making claims that don’t — their track record is horrible. Their track record is literally horrible.
Wiser, more mature, more evidence-based.