I just want discussions on it to be more nuanced than "AI nothing but a grift like NFTs" or "AI is going to transcend reality and fly off into space" polarities of extremes. I don't even mind if people come to the conclusion of seeing it as a net bad, as long as they're doing proper analysis of what it is in practice.

I maintain that a lot of the angst is due to the lion's share of the conversation about it (in the west) being decided by the investor class / marketing teams under capitalism and the changes they make in the world, which gives people good reason to dislike it. Like I'm pretty sure there's been data that people in China tend to have a more positive view of AI than in the west. But as a concept, what commonly gets called AI now (which is mainly gen AI in western conversation about it) is primarily a new form of automation. The singularity stuff is kind of a pipe dream and not needed for automation to reshape the world, which was happening long before AI became the big buzzword. What makes gen AI more distinctive is, in particular, the fundamental design of language models to mimic human conversation as accurately as possible. Which introduces unique questions about the implications of having such a convincing virtual imitation machine that more obviously disembodied forms of automation wouldn't cause. But underneath it, it is still automation, not an authentic recreation of a person. However, it's good enough at imitating and following along that some people get taken in by the simulation of it, which can create new problems or worsen existing ones, especially in a culture like the US's that is already highly immersed in bullshit. On the other hand, some people have found the simulation of it to be helping for processing difficult things in a context that feels safer than what they otherwise have available.

So it's complicated.