Raw LLM Responses

Well, he’s a professor researching the area of AI safety at an institute that researches and publicises the subject of AI safety. I’m not sure “crackpot” is the best description (unless you have a doctorate I’m unaware of and some secret knowledge on the matter). I’m also moderately curious as to how talking about it “muddles the water”. Granted, the guy is pushing the book he’s written, and the article about it is clickbaity. However, his opinion is just an argument, that can be evaluated on its own merits. If you’re curious as to *why* he’s making this argument, then it’s because it’s based on logical conclusions from computer science (experiments on superhuman AIs being scarce, and not a good idea).