ChatGPT has recently gotten surprisingly good at reasoning out problems.
I like to throw little engineering challenges to it to see what it will do - as of a couple days ago it's changed.
1. It's now saying "I don't know enough to say what this is and I don't want to hallucinate"
2. It's doing really good logical deduction, now finding ways of looking at a problem that I didn't think of (and which once you see them they do add a lot of clarity)
I like to throw little engineering challenges to it to see what it will do - as of a couple days ago it's changed.
1. It's now saying "I don't know enough to say what this is and I don't want to hallucinate"
2. It's doing really good logical deduction, now finding ways of looking at a problem that I didn't think of (and which once you see them they do add a lot of clarity)
