> LLMs are mere text generators. Trained on unimaginable amounts of text, they string together words in coherent sentences based on statistical probability of w

> LLMs are mere text generators. Trained on unimaginable amounts of text, they string together words in coherent sentences based on statistical probability of words following other words. But they are not "intelligent" in any real way — they are just automated calculators that spit out words. These kind of takes are getting tiring fast. I’d even go as far as to qualify them as midwit takes. Yes you can reduce LLMs to their output channels, or trivialise what their doing by using words as « statistical ». Facts: ChatGPT, a rough demo based on a model that is already obsolete, is nearly indistinguishable from the average college student.