Do you have a dumb question that you're kind of embarrassed to ask in the main thread? Is there something you're just not sure about?
This is your opportunity to ask questions. No question too simple or too silly.
Culture war topics are accepted, and proposals for a better intro post are appreciated.
Jump in the discussion.
No email address required.
Notes -
How do you all interact with LLMs?
I’ve seen a few articles recently noting the rise of AI as a buddy / therapist or whatever. It’s usually beside the point of the article but an implicit notion is that lot of folks regularly ‘chat’ with AI as if it were a person.
Which I find baffling. Outside of the very early novelty, I find this way of interacting extremely boring and tedious, and generally find the fact that AI wants to get conversational with me a general frustrater.
If I’m not using AI as a utility ‘write X, troubleshoot Y, give me steps for Z’, and I’m using it recreationally / casually, it’s more akin to web surfing or browsing Wikipedia than chatting on a forum or whatever. I will use it as an open format encyclopedia and explicitly not as a conversationalist sounding board. And i genuinely find negative value in the fact that the former is constantly interrupted with the attempt to be the latter.
So my question is again, how far outside of the grain am I?
OK, this has mystified me for a long time. I use LLMs for various editing, writing, coding tasks, occasionally to kludge a moderator for party games, to simulate human feedback on human-oriented questions, and once in a long while to suggest a starting point for a lit review or to locate a half-remembered link. But can you help me understand the "encyclopedia" and "web surfing replacement" use-cases, when we have actual encyclopedias and a web to surf?
When I see a granny or a teen just asking ChatGPT, I assume it's because they can't internet, can't read, or don't give a shit about the quality and provenance of their information, but for a super-online, epistemically hyper-aware Mottizen to do this feels like hearing someone say they hire a guy to order all their food, chew it and spit it in their mouth.
When looking for answers to programming questions, lots punctuation gets stripped out of search queries, and many language keywords are stopwords that don't get included in a search. But to an LLM, they're more tokens.
Another thing I've found useful is to get one to surface general issues in first-pass troubleshooting and then go look for actual forum threads documenting those issues. This helps you find where the experts are and cross-check the output against a real discussion.
Yeah, I do understand using the LLM for search or even for a link-enriched overview to cross-check with real resources, as you describe.
I mostly get confused when people Ask ChatGPT, consume the generated content and stop there, which (for a Motte level of understanding "assertions can be wrong," "sources can be mistaken," "context matters," "models sometimes confidently hallucinate") seems like a weird combination of definitely caring and definitely not caring about whatever fact you're researching.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link