I asked it about turing completeness in bitcoin, and how in…

I asked it about turing completeness in bitcoin, and how in 2015 Craig talked about how it was. It came back with a lengthy post about how it was impossible due to disabled opcodes, and script size limit, and tx size limit.

When I then asked if it could be done in original form of bitcoin in2009, it basically said yes, but it 'practically' was not because... "Block size limits, script size restrictions, transaction fees, and the absence of native loops made it infeasible to run arbitrary computations."

Transaction fees in 2009 were 0... lol.

Anyway, I then discussed how the game of life could run on bitcoin, and it said it could only be run on bitcoin sv.

I then asked a few other questions, which were equally hallucinated and partially correct answers.

Anyway... I then went back and re-read your post, and have completed the task required.

So... it didn't bring up the court case, or his satoshi claims in this response, but it did bring it up here and there during my previous discussions.

It is definitely is on the non smart side of the equation.

As I see 'smart' and 'intelligent' as 2 different things.

One can sense a pattern, then dig into it to find the truth. The other is just regurgitating facts and info.

I have the same issue with google's AI answers it gives...

Half the time you ask for something specific, that is to not include this particular term. It will then spew out a number of hallucinated answers, often included the thing you said not to include. It's terrible...

At least grok is a little more in depth, and a little more accurate, but it lacks the 'smart' that truly makes an AI, actually what people want and expect.

Grok just regurgitates. It doesn't dig for truth. :/

Oh, $metal is my handcash. Thanks :) Hope my reply helps you. It certainly informed me about how unsmart grok is, and confirmed once again that AI may never get past the hallucinating answers group think stage, and truly become smart.