People who have a more in-the-middle opinion generally don’t talk about AI a lot. People with the most extreme opinions on something tend to be the most vocal about them.
Personally I think it’s a neat technology, and there probably exist use-cases where it will work decently well. I don’t think it’ll be able to do everything and anything that the AI companies are promising right now, but there are certainly some tasks where an AI tool could help increase efficiency.
There are also issues with the way the companies behind the Large Language Models are sourcing their training data, but that is not an inherent issue of the technology. It’s more an issue with incorrectly licensing the material.