Doesn't that somewhat miss the point? The models can't see and check their work, for instance.

I'm not clear what you are saying. I am curious about what humans can do in the same context. It's becoming a standard I think for some tasks. E.g. how does a Gen AI stack up against a human in quality and "cost". Perhaps I'm missing what you are asking.

They can. Ask them to generate SVG , call a renderer and let them inspect the bitmap/png. They can learn about it.

They can, but for this specifc test, they don't. It's a oneshot with no feedback.

Technically you would pay programmers to write SVG code with no visual feedback if you want a straight 1:1 comparison.

Kinda. It's the difference between measuring the end product vs how it's made.