It would be fine if there was a big warning "will make mistakes!" but many people seem to believe it's flawless output.

I think they are right about this:"In general, users still need to closely monitor LLM systems as they operate and complete tasks on their behalf."

@madeindex We *really* want these things to be autonomous but they’re not there yet.

But I think they will be within 12 months - not just big company models, but the ones you run at home too. Not sure what that means for us. I do think some jobs will disappear, but a lot of new jobs and opportunities will appear.

That would be great, I do wonder however if they will ever get there.

Somebody sent me this recently, I didn't entirely get it, but it seems like there are structural issues: nature.com/articles/s41586-024…

I sincerely hope #AI will empower humanity to do other, more useful things (especially as it reduces digital work), and even reach the next stage of advancement, don't trust those corporations with it though, fingers crossed eh :)

AI models collapse when trained on recursively generated data - Nature

 Analysis shows that indiscriminately training generative artificial intelligence on real and generated content, usually done by scraping data from the Internet, can lead to a collapse in the ability of the models to generate diverse high-quality out…

^Nature