"The Singularity"
|
|
The concept of "The Singularity", popularized by inventor and futurist Ray Kurzweil in the mid-2000s, refers to a point in the (not-so-distant) future when the intellectual capacity of computers exceeds that of humans. This idea has long existed as a negative force in the science fiction: such as the murderous computer HAL in 2001: A Space Odyssey or the malignant artificial neural network Skynet in the Terminator movie franchise. In the words of Stephen Hawking:
The development of full artificial intelligence could spell the end of the human race … It would take off on its own, and re-design itself at an ever-increasing rate. Humans, who are limited by slow biological evolution, couldn’t compete, and would be superseded.
Even before the full Singularity occurs, some experts have expressed their doom about radiology in particular. In 2016 the famous computer scientist Geoff Hinton stated in a public lecture:
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 get a lot more experience. It might be ten years but we’ve got plenty of radiologists already. I said this at a hospital and it didn’t go down too well.
In the months after making this provocative statement, however, Hinton became aware of the many important non-interpretive functions performed by radiologists and substantially modified his view. Today, 6 years later, radiologists have not yet become obsolete but are clearly more focused on the impact AI will have on their future practice of medicine.
As for me, I am happy to be mostly retired from clinical radiology; I need not worry a computer is going to take over my job as a radiologist. And certainly not as a grandfather. I am proud to have developed a considerable knowledge base in radiology during my career, and through my skills was able to help a large number of patients, including diagnosing a number of rare diseases missed by others. I enjoyed tremendously passing on what knowledge I had gained to several generations of radiology trainees and now to an even wider audience through this web site.
I can't predict if or when the Singularity may occur. But whether or not that point is ultimately reached, it is important to understand how our jobs and and lives will be dramatically transformed by AI in ways we cannot fully imagine.
Concerning radiology in particular, I have no doubt that within the next decade, computers will be able to read imaging studies with greater speed and accuracy than the best radiologists. But before we bemoan this fate, let us never forget the potential good this will bring to humanity. Computers using AI will discover patterns of disease we humans have long overlooked. They will be able to identify genetic subtypes of disease that can alter therapy. They will be able to precisely calculate small changes in lesion size and biological response. They will diagnose the rarest of diseases, previously the provenance of super-specialists. And they will never miss common but subtle abnormalities: a tiny lung nodule, hairline fracture, or early cancer — unlike we bleary-eyed humans might at the end of our long shifts.
Concerning radiology in particular, I have no doubt that within the next decade, computers will be able to read imaging studies with greater speed and accuracy than the best radiologists. But before we bemoan this fate, let us never forget the potential good this will bring to humanity. Computers using AI will discover patterns of disease we humans have long overlooked. They will be able to identify genetic subtypes of disease that can alter therapy. They will be able to precisely calculate small changes in lesion size and biological response. They will diagnose the rarest of diseases, previously the provenance of super-specialists. And they will never miss common but subtle abnormalities: a tiny lung nodule, hairline fracture, or early cancer — unlike we bleary-eyed humans might at the end of our long shifts.
So, my personal advice is that you must focus your career on being more than just a "film reader". (You'll still be OK for a few years if that's all you want to be. Perhaps you'll survive for a decade, but the machines will ultimately beat you at this game). Think about what you do that will be hard for computers to replicate — like comforting a patient when you've made a diagnosis of cancer; advising a mother on the best interventional procedure for her child; consulting with other physicians about imaging strategies for their patients; treating and counseling a patient who developed hives after contrast administration.
To me at least, the best strategy for surviving in the face of a potential Singularity is simply to focus on being human.
To me at least, the best strategy for surviving in the face of a potential Singularity is simply to focus on being human.
References
Kurzweil R. The Singularity is Near. New York: Viking Books, 2005. [Downloadable from this link]
Vinge V. (1993). Vernor Vinge on the Singularity. San Diego State University, 1993. [Downloaded from http://mindstalk.net/vinge/vinge-sing.html 17 Jan 2022] (First description of the "Singularity" as applied to the ascendency of AI over humans, though it had been used by von Neumann in a more general context in the 1950s.)
Kurzweil R. The Singularity is Near. New York: Viking Books, 2005. [Downloadable from this link]
Vinge V. (1993). Vernor Vinge on the Singularity. San Diego State University, 1993. [Downloaded from http://mindstalk.net/vinge/vinge-sing.html 17 Jan 2022] (First description of the "Singularity" as applied to the ascendency of AI over humans, though it had been used by von Neumann in a more general context in the 1950s.)
Related Questions
What is artificial intelligence (AI)?
What is artificial intelligence (AI)?