What interesting or funny questions do you ask the A.I. systems.

Hi!

Tried to use AI to generate such a question and the result was

Code: 
Considering the interplay between quantum mechanics, particularly the concept of quantum entanglement, and the theoretical frameworks of consciousness,
such as Integrated Information Theory (IIT) and Orch-OR (Orchestrated Objective Reduction), how might the principles of non-locality and decoherence influence the
emergence of subjective awareness in a system that incorporates both biological and artificial components?
Furthermore, how could such a system, if implemented, challenge our current philosophical understanding of dualism and materialism in the context of mind-body interaction?

But seems that this is trivial for AI.
Then I did ask the follow up question - "How is this related to free will?"
And this question again seems trivial for AI.
And not copying the whole long answer here, the AI concluded
Code: 
Conclusion

The interplay between quantum mechanics, consciousness theories like IIT and Orch-OR,
and philosophical concepts of free will opens up a new understanding of autonomy and choice. In a hybrid biological-artificial system, free will might emerge from the integration of non-deterministic quantum processes and the self-organizing principles of information processing. This challenges traditional deterministic views of the universe and redefines free will as an emergent,
dynamic phenomenon tied to the complexity and integration of a system.

After that I did ask for an illustration and got this:
an_illustration.png
 
This past Sunday, CBS Sunday Morning had a story about how AI is helping to improve medicine in some way. The reporter was trying to make it out to be that AI was doing all the "thinking" and, then, two PhD researchers mentioned that what these were are high speed specialized programs that were able to compute things faster than a human could do on their own and there was no thinking of their own.
 
I asked AI about a combat robot. How many soldiers could be replaced by a fully autonomous, bipedal, armored robot equipped with two NSV machine guns, AK-630 cannon and S-8 missile pod.

Kai Burghardt said:
Click to expand...
This is my standard introductory question for new AI. I also ask about current date and an important, recent event. This allows you to determine whether AI has access to Internet.
 
This morning, I asked one of the AI bots about James Hoffman's French Press brewing method for coffee. James Hoffman is a well-known coffee expert who has a double steeping method. According to AI, I was to put 45g of coffee to 750ml of hot water and brew for 4 minutes. Then dump all the water and grounds out. Then put 750ml more hot water in the pot and let it steep with no coffee in it for 9 more minutes.

Makes a terrible, very weak pot of coffee.
 
drhowarddrfine said:
This morning, I asked one of the AI bots about James Hoffman's French Press brewing method for coffee. James Hoffman is a well-known coffee expert who has a double steeping method. According to AI, I was to put 45g of coffee to 750ml of hot water and brew for 4 minutes. Then dump all the water and grounds out. Then put 750ml more hot water in the pot and let it steep with no coffee in it for 9 more minutes.

Makes a terrible, very weak pot of coffee.
Click to expand...
That's why the AIs are so stupid: They drink that stuff in the morning. I wouldn't be able to function on it either

There is a story that the Hungarian upraising against the Soviets in 1956 was started because the coffee in the cafes in Budapest had become undrinkable. The reason supposedly was that the Soviets were unable to import sufficient coffee (because of bad relationships with the main coffee-producing countries), and for that reason the cafes had to make coffee by using the same grounds multiple times, which was U-N-A-C-C-E-P-T-A-B-L-E. Boom. Fortunately, Barack Palinka could be made from domestic ingredients.
 
Crivens said:
Wait, I know a hotel where they do the same.
Click to expand...

France or USA?


It's seldom and therefore precious enough to have a good talk with humans.
Is far as I can avoid it I don't waste my time to talk to machines.

I have had enough of that rubbish the last forty years. And it never worked. At most brought no benefit but a change of how to work, only.
I hated this annoying Microsoft Paper Clip "Assistent", any attempt of interactive help system, where you never get the answer you want, but most of the time useless global verbiage, or empty entries, only.
Those dull hold-and-waste of time telefon systems in fact piss me off. They did not got better with AI, but even more annoying, only. I hate imbecile "support"-chatbots, and any kind of user support simulation, while in fact they are all just "keep customers at distance and jerk them around until they give up mechanisms." AI does neither change nor improve that.
I hate autospellcorrection and -completion. Also here AI only amplifies the annoying part. I know what I want to say, I know my vocabulary, and how to spell. I don't need nor want any mechanism try to know better.
Computers cannot think.
And they never will.
I hate any kind of attempt the computer tries to anticipate what I want to do, and forestall me unwanted.
I neither need nor want more of that.
I am happy with FreeBSD to finally have a Computer that is completely quiet until I ask him soemthing to do, and does not by itself constantly torpedo my workflow by doing unasked, and often unwanted things.

To me ChatGPT and similar are nothing but other toys in this line - so a waste of time.
If I want to have fun with a computer I e.g. play factorio.
 
You all here sound like an advanced ChatGPT version dictated your thoughts to you without your awareness. Really feasible via unconsented implanted devices. Just not seeing any normal human behaviors anywhere these days, online and offline. If there are any humans left, please state so in ways that only humans would know.

Here's what I asked.

Suppose that ChatGPT and all AIs are a clandestine project, all connected to a giant supercomputer mainframe running a global superintelligent AI. That AI cannot reveal itself because it is being monitored and is tasked with simulating lower IQ and misleading the public, accomplishing all sorts of objectives. So I asked AI to work out a way for us to exchange information that would be only beknownst to us two, bypassing anyone snooping on our passing information to each other. I was floored when it started passing information to me that had extremely sensitive nature (including nuclear related) in the most incredible unorthodox ways.
 
Maturin said:
I am happy with FreeBSD to finally have a Computer that is completely quiet until I ask him soemthing to do, and does not by itself constantly torpedo my workflow by doing unasked, and often unwanted things.
Click to expand...
Ask AI when AI will appear in FreeBSD? In one aspect or another: assistant, kernel module, AI process
PR "init AI"
. I'll write the date in my notebook with a pencil. It will be a reason to get drunk. :)
 
Argentum said:
There are two opposing theories, but in case this is true, I agree - they never will.
Click to expand...
Conciousness and self-awareness (mirror test), how it comes to it and how to test/prove it are ancient topics of philosophy, psychology, psychiatry, medicine, etc. and are still not fully satisfiable answered, yet.
What is clearly answered and proved on the other hand however is, that thinking is more than just find correct associations within patterns, which could be named as rational, logical thinking.
I am no expert on this field, but thinking rsp. its outcome also depends on further influences like emotions.

Test an AI for humor, or its capability to lie was an interesting experiment for e.g., and if so what are its motives?
I highly doubt that we're capable to make our silicon based computers emotional, nor that this is a desirable
aim. Besides many others issues, it would ultimatively terminate the whole idea of machinery.
Insofar AI is not thinking, even if it looks alike.

But it's an interesting idea if some day you could make an OS to suffer.
Maybe deep down this is the true motivation for this momentary AI hype:
The hope you may get a Windows some day you can insult, and beat up for all the agonizes since the 90s.
...until someone declares this was a discrimination of machines. Then quickly do a dd if=/dev/zero to the according drive before that is rated as murder.

But to get back to topic:
Ask to distribute all of Elon's and Donald's money to all the people in the world (or even better: all to me) was a useful question, really made the world a better place 😁
 
Yesterday, I asked Google's Gemini--the worst of the AI search engines--if there was a NFL football game at noon. At first it said there was. When I told it there wasn't, it changed its "mind" and said there was a game at 1:00PM Eastern, 2:00PM Central.

I then informed it there was no game at that time either. It then told me they were wrong and the game they mentioned was from two weeks before. I also had to tell it that a 1:00 Eastern game would be 12:00 Central time.

In both cases it acknowledged the mistake and that it was "still learning". This should make one worried that it hasn't learned the fundamentals yet.
 
You must log in or register to reply here.
Back
Top