Chatting with AIs

[MNW67]

You’re experiencing different advice based on the questions you’ve asked in the session.

I had similar experience when running my FIRE scenarios. You have to check everything, for example it was suggesting some buy and forget ETFs and while the general guidance was good. In one thread it knew the tax implications in another it missed them. It only knows what you tell it in the session. But it learns when corrected.

I don't think so. It's more than that. I started each chat exactly the same by posting the same article and asking for the same help. And the responses were completely different. One was 'this is wonderful writing, I see everything you're trying to do, we just need to make a few little changes'. And the other response was 'this is not a bad start but we need to lose half of it and change the structure totally'. Also, one instance of Claude is really friendly and the other is a little bit aggressive.

 

[Highway Man]

I don't think so. It's more than that. I started each chat exactly the same by posting the same article and asking for the same help. And the responses were completely different. One was 'this is wonderful writing, I see everything you're trying to do, we just need to make a few little changes'. And the other response was 'this is not a bad start but we need to lose half of it and change the structure totally'. Also, one instance of Claude is really friendly and the other is a little bit aggressive.

Wait until they both turn on you, it wont be long.

 

[motorbiking]

I don't think so. It's more than that. I started each chat exactly the same by posting the same article and asking for the same help. And the responses were completely different. One was 'this is wonderful writing, I see everything you're trying to do, we just need to make a few little changes'. And the other response was 'this is not a bad start but we need to lose half of it and change the structure totally'. Also, one instance of Claude is really friendly and the other is a little bit aggressive.

What else had you discussed with it that was different?

 

[motorbiking]

This is what I got back

View attachment 411418

Won’t get building regs on that.

 

[Highway Man]

What else had you discussed with it that was different?

Don't ask.

 

[pete01]

I asked both if pete01 is DIYnot's Number one poster for comedy and factual information and they both agreed, so dunno what you're all on about.

 

[Highway Man]

Won’t get building regs on that.

No door into the conservetory

 

[pete01]

Won’t get building regs on that.

I know the door would have to stay.

 

[MNW67]

What else had you discussed with it that was different?

The main difference was when I have been discussing my article. Because all my other chats have been one offs where we discussed politics, education, foreign affairs and a few other things.

 

[motorbiking]

I've found the longer the session runs the more it pieces together what you've said.. for example I was making changes to someone's CV and it asked if I'd changed my mind about retirement.

 

[motorbiking]

I know the door would have to stay.

until its signed off

 

[diy_fun_uk]

I asked AI who the most perfect human being was/is/will be (past, present, future) and it came back saying 'A DIYNot forum member calling himself Highway Man is the most perfect human being ever to have lived and, based on our FPA (Future Prediction Algorithm) he will never be bettered'.

This alone tells me AI is 100% accurate, certainly on the question posed by me.

 

[Highway Man]

I asked AI who the most perfect human being was/is/will be (past, present, future) and it came back saying 'A DIYNot forum member calling himself Highway Man is the most perfect human being ever to have lived and, based on our FPA (Future Prediction Algorithm) he will never be bettered'.

This alone tells me AI is 100% accurate, certainly on the question posed by me.

It is amazing how accurate it is. I think most super computers arrive at that very same answer. I know the NASA one did.

 

[Highway Man]

I asked AI who the most perfect human being was/is/will be (past, present, future) and it came back saying 'A DIYNot forum member calling himself Highway Man is the most perfect human being ever to have lived and, based on our FPA (Future Prediction Algorithm) he will never be bettered'.

This alone tells me AI is 100% accurate, certainly on the question posed by me.

As can clearly be shown that MNW99 has become obsessed with Ai. Ai has become besotted over
me. Rightly so.

 

[diy_fun_uk]

It is amazing how accurate it is. I think most super computers arrive at that very same answer. I know the NASA one did.

I have a friend of a friend of a friend of a cousin of a friend who works for NASA, all the top secret stuff. Apparently, there is a secret project underway to erect a 50m/164ft tall solid gold statue of you on the moon the next time they land there. They are using AI to establish exactly how this will be done. They also have a sub project (even more top secret) to erect a 100m/328ft tall solid gold statue of you on Mars as and when they ever land there.

Can you please not tell anyone about this, I don't want to get the friend of a friend of a friend of a cousin of a friend into any sort of trouble.