I'd argue that his point is worse than this.

He's not comparing them to humans, he attributes knowledge/understanding (enough to pass his bar for "is AI") to yolov5 models, xgboost trained trees and as far as I can tell closed source transformer based models too. But not ChatGPT.