I’ve been watching AI generated videos. They’re freaky and hilarious. The AI obviously does not really understand what a face or a hand is, it just calculates statistically what it thinks these things should look like. As Daniel Dennet says, you can sometimes have competence without comprehension. I’m not sure how far this can go before some real comprehension is required. How will we get machines to have "real" comprehension? I guess we’ll have to get a better idea of what that is before we can get the machines to do it.
From what I’ve been reading about animal studies, it seems there has to be an innate expectation of the world pre-written into the mind. They say rats are dreaming about the world before they ever open their eyes and go out and experience it. When they see it, it’s what they’re expecting, with the specific details filled in. So to "comprehend" the situation, the AI needs to be able to look at a picture or video and make some comparison of what’s contained in that data to what it expects the real world to be like. I get the impression that this is extremely complicated. Those who want to reduce consciousness to a simple equation or written statement that can be put on a T-shirt I think are not being realistic. I'm supposing here that consciousness and comprehension of the world and self are closely related.