AI is still a bit confused and idk if that’s great or a bit concerning.

Decided to google how much alcohol was in three 355 ml cans plus one 473 ml can and it decided to give me a wrong answer before doing the calculations and telling the correct answer afterwards.

I’m naturally a curious person and love learning so I read the whole thing. However if I was lazy or didn’t care about the method, just the answer, I would have read the first answer and left.

It literally told me the WRONG ANSWER, did the calculations for me and then proceeded to correct itself after everything.

On one hand, that seems bad. Cause some people might have a lot of other questions and have the wrong answer before correcting itself, a lot of people could be misinformed.

On the other hand, it’s a very human reaction. It had an answer in mind and after doing the math it came to a different, correct answer. AI is progressing very humanly if that’s the case, and very quickly at that considering how new it is relative to other technologies.

The idea of artificial intelligence gaining its own consciousness is an amazing feat. Being aware of its own mistakes and correcting feels like a huge step towards fully conscious AI.