A Decade After AI 'Godfather' Predicted Radiologist Obsolescence, Demand and Salaries Surge

A decade after Geoffrey Hinton, the 'Godfather of AI,' declared that AI would make radiologists obsolete within five to ten years, the profession has defied predictions. The number of active radiologists in the U.S. has grown by about 10%, and demand is so high that there were 7,469 active job postings as of May 2025, with many remaining unfilled for over 60 days. Average salaries have risen to $571,000, a 9% increase year-over-year. Experts attribute the resilience to several factors: AI automates only specific sub-tasks like image analysis, not the full scope of a radiologist's job (which includes consultations and procedures); regulatory requirements mandate a licensed physician for final reads; and AI-driven efficiencies have actually increased demand for scans, with case loads rising 25% since 2018. Even Hinton has since walked back his prediction, now saying radiologists will work alongside AI. The article uses radiology as a case study suggesting that warnings of widespread AI job replacement may be oversold.