If that's the case, then as long as the "AI research" was thoroughly checked by a human, then I can't imagine how he'd have even been caught using AI. So, it's strange that he was caught.

I just did a search on it and apparently someone said that a particular sentence in the video sounded like it was written by AI. He said that it wasn’t written by AI and in fact wasn’t written at all as it was ad-libbed during filming. He then confessed that he has been relying too much on Chat GPT for research and that his script did sound too much like AI.

Kotaku article

It was a very niche topic about crocheting or something, I think? The problem is that he didn't know about it, so he was researching it as part of the script. And because he didn't know about it, he didn't realize that some of the information was wrong. Well, the crochet community knew it was wrong, because of course they'd know more about the topic. So that got people questioning how Hank could have come to this wrong information which led to people further dissecting the video and concluding that it was AI assisted.

So, Hank, a human, looks to learn about a topic he wants to talk about it. He, a human, does check the material, but he's not really qualified to know if the material is right (hence wanting to learn about it).

I think what he should do (and I think he kind of touched on this, but I don't know where) is he should reach out to experts in that field to verify his scripts. Rather than just trying to verify them himself. Especially if its about a topic he's not super confident in.

But most importantly... none of that makes the man out to be some kind mega villain, AI-slop loving, earth killing hypocrite that people are making him out to be. Just a dude who wanted to talk about something and made a mistake. And yet people want to run him through some kind of humiliation ritual and penance. It's sad.