The Works in Progress newsletter had a thoughtful article by Deena Mousa on AI and radiologists.
Radiology accounts for the vast majority of AI medical devices cleared for use. As advances in AI showed positive progress in studying scans, Geoffrey Hinton – Turing Award winner and one of the fathers of the modern AI wave – declared in 2016 that ‘people should stop training radiologists now’.
However, the opposite has happened.
“In 2025, American diagnostic radiology residency programs offered a record 1,208 positions across all radiology specialties, a four percent increase from 2024, and the field’s vacancy rates are at all-time highs. In 2025, radiology was the second-highest-paid medical specialty in the country, with an average income of $520,000, over 48 percent higher than the average salary in 2015.”
As Deena explains, there are three things that explain this.
First, while models beat humans on benchmarks, the standardized tests designed to measure AI performance, they struggle to replicate this performance in hospital conditions. Most tools can only diagnose abnormalities that are common in training data, and models often don’t work as well outside of their test conditions.
Second, attempts to give models more tasks have run into legal hurdles: regulators and medical insurers so far are reluctant to approve or cover fully autonomous radiology models.
Third, even when they do diagnose accurately, models replace only a small share of a radiologist’s job. Human radiologists spend a minority of their time on diagnostics and the majority on other activities, like talking to patients and fellow clinicians.
She also calls out “Jevon’s paradox” – the cheaper something becomes, the more likely we are to use it.
In many jobs, tasks are diverse, stakes are high, and demand is elastic. When this is the case, we should expect software to initially lead to more human work, not less. The lesson from a decade of radiology models is neither optimism about increased output nor dread about replacement. Models can lift productivity, but their implementation depends on behavior, institutions and incentives. For now, the paradox has held: the better the machines, the busier radiologists have become.
She makes a beautiful point as she extrapolates her lessons learnt from this.
Artificial intelligence is rapidly spreading across the economy and society. But radiology shows us that it will not necessarily dominate every field in its first years of diffusion — at least until these common hurdles are overcome. Exploiting all of its benefits will involve adapting it to society, and society’s rules to it.
Indeed.
