It goes for Large Language Models as it goes for everything else: Garbage in, garbage out. This article and its author really misses the point. > This is just

It goes for Large Language Models as it goes for everything else: Garbage in, garbage out. This article and its author really misses the point. > This is just obviously not even a plausible simulacrum of a conversation. It is not supposed to be. It is supposed to be an interactive search engine. Bing with a layer of GPT on top or however you would like to abstract it. It is trained, like chat-gpt is, to give responses in a particular structure and with a strict ruleset of what it should not allow the user to request etc. I enjoy reading, but this article - Sorry to say its just messy and tries too hard to up the Lix score.