Worried That AI Might Make You Obsolete?

Key Points:

• Key assumptions overlooked diagnosis' statistical nature which requires both machine and human strengths. AI and radiologists balance sensitivity and specificity.
• AI bifurcated work into prediction and judgment steps rather than wholesale replacement. This increased human judgment's value.
• Practical inconsistencies between experimental and real-world AI performance persist, hampering adoption. Humans readily transfer contextual skills.
• Technologists focused narrowly on automating tasks rather than radiology's multifaceted whole involving fluid human-AI collaboration.
• Job disruption is filled with uncertainty. Work comprises complex social systems, not just replaceable discrete tasks. Reciprocal integration shapes futures more than automation.
• Predictions require accounting for both automation capabilities and human adaptability within wider systems. False dichotomies between full automation versus none are misleading.
• Trends like increased demand for radiologists post-AI integration exceeded expectations of obsolescence, showing reality is more nuanced than headlines suggest.
• Making better predictions about how work changes with AI requires a deeper understanding of the integrated and complex nature of decision making: prediction, judgment, action in complex human systems.
• We find three laws—abundance, complexity, and agency—which combine to create new decision making systems with AI.

Flawed Predictions

In 2016, Geoffrey Hinton, a pioneer in deep learning, claimed, "We should stop training radiologists now, it's just completely obvious within five years deep learning is going to do better than radiologists". Similarly, in 2017, Vinod Khosla, a prominent venture capitalist, asserted that "the role of the radiologist will be obsolete in five years".

Radiology, it seemed, was destined for obsolescence, with artificial intelligence taking over the reins by 2020. Machines, according to Oxford economists, would replace doctors as many tasks within professional work were deemed routine and process-based, lacking the necessity for judgment, creativity, or empathy.

Yet, these predictions from technology visionaries failed to materialize. What led to their glaring misjudgment? And, more broadly, what can we learn about AI-driven human obsolescence?