Quote:
Originally Posted by TorqueDog
Copilot has been good for the soft-ball tasks I've thrown at it (particularly since it can leverage corporation documentation).
However, I personally pay for ChatGPT Plus and it just did the same thing to me as it sounds like it did to you. I fed it three documents to use as its foundational basis for reviewing a fourth document, and it started making up clauses in the fourth document and flagged them as violations....
|
Out of curiosity, did you try this with o3 (base) or 4.1? I only ask because o3 seems far better at the more complicated high-stakes workflows and analysis, and 4.1 has a 1 million token context which could handle your documents better.
Apparently o3 pro (available at that $200/month tier) is blowing some pants off, but I don't have the money to try it out.
Quote:
Originally Posted by Fuzz
https://www.extremetech.com/computin...-an-atari-2600
LOL. I love that it tried to make a human-like excuse instead of owning it's sucking. Wait, I was about to joke about how audacious it is to call it "AI" if it can't play chess, but perhaps it's doing a much better job at emulating emotional responses than thinking. Which would be interesting if we made an emotional bot before an intelligent one. |
These are always very funny comparisons, and frankly anything that keeps people out of the AI pool so that I can continue to play is all thumbs-up to me. But it's not really a logical comparison. It's a bit like saying because ChatGPT can't count the number of R's in "Strawberry" it's not as useful as a dictionary. They're different things and they don't operate the same.