I’m always sceptical about results like these. I was told that waterfall always failed when I’d worked on successful waterfall projects with no fails. The complaints about waterfall were exaggerated as I think are complaints about agile. The loudest complaints seem to always be motivated by people trying to sell sonething
Why would anyone expect “nuance” from a generative AI? It doesn’t have nuance, it’s not an AGI, it doesn’t have EQ or sociological knowledge. This is like that complaint about LLMs being “warlike” when they were quizzed about military scenarios. It’s like getting upset that the clunking of your photocopier clashes with the peaceful picture you asked it to copy
I find this extraordinarily unconvincing. Firstly it’s based on the idea that random graphs are a great model for LLMs because they share a single superficial similarity. That’s not science, that’s poetry. Secondly, the researchers completely misunderstand how LLMs work. The assertion that a sentence could not have appeared in the training set does not prove anything. That’s expected behaviour. “stochastic parrot” wasn’t supposed to mean that it only regurgitates text that it’s already seen, rather that the text is a statistically plausible response to the input text based on very high dimensional feature vectors. Those features definitely could relate to what we think of as meaning or concepts, but they’re meaning or concepts that were inherent in the training material.
Yeah agree with you about indie games. Still some genuine passion there