We Test the AI. How Do We Test the Oversight?

As the global AI community gathers in Amsterdam for World Summit AI, one question is becoming increasingly important in healthcare: we test the AI, but how do we test the oversight?

Putting a human in the loop sounds reassuring. But human involvement alone doesn’t automatically make an AI system safer.

If an AI flags something on a medical scan and a clinician disagrees, who is right? Has the clinician caught an AI error, or has the AI detected something the clinician missed?

Effective oversight depends on whether people have the information, tools and authority to challenge AI — and whether there is evidence that their involvement actually improves outcomes.

It leads to a simple but important question:

We demand evidence that the AI works. Shouldn’t we also demand evidence that the oversight works?

This Thursday at World Summit AI, Helia Mohammadi of City of Hope and Dustin Raney of SuperTruth will explore exactly that during “AI in healthcare: How safe is safe enough?”

📅 Thursday 8 October
⏰ 15:30
📍 Health and Humanity Stage, World Summit AI 2026

InspiredMinds! Community Hub

Inside the InspiredMinds! Community Hub, you’ll find deeper insights, expert perspectives, and practical discussions from the people building and deploying AI across Europe and beyond.