Hmm, thought I’d try something out and looks like maybe it does actually work to some degree?
Basically, adding “If you don’t know, you don’t have to make something up, just say ‘I don’t know’” to the end of an LLM prompt to try and cut down on the bullshit (doesn’t fix the environmental footprint, though).
Background on the watch question: afaik, there are no LED watches with automatic movements, although Hamilton has one with an LCD display.
#AI #LLM #IDontKnow #hallucinations #bullshit
Aral Balkan
in reply to Aral Balkan • • •Ben Ramsey
in reply to Aral Balkan • • •It doesn’t accept instructions. It’s only responding to the input with the most probable output, which happens to be “I don’t know,” when you prompt it to say “I don’t know,” if it doesn’t know.
This is why the “ignore all previous instructions” meme is silly. 😂
Aral Balkan
in reply to Ben Ramsey • • •Hyperlink Your Heart
in reply to Aral Balkan • • •Aral Balkan
in reply to Hyperlink Your Heart • • •Andres Jalinton
in reply to Aral Balkan • • •static1.squarespace.com/static…
Aral Balkan
in reply to Andres Jalinton • • •Aaron Caskey-Demaret
in reply to Aral Balkan • • •I feel like, as a word prediction engine, it would then reply "I don't know" to things that it would otherwise give correct answers to.
... and that's fine. I'd rather it do that than give people advice that'll kill them But then it'll probably still do that too sometimes.
Dany
in reply to Aral Balkan • • •Denis 🌏
in reply to Aral Balkan • • •I wonder if one could also add to the prompt "and give me your confidence in your answer, high, medium, or low" and if that would provide anything useful.
(but again, doesn't help the environmental footprint. I hate that I find these tools actually useful in helping me work more effectively.)
#LLM