Author Topic: Sentient AI or?  (Read 10855 times)

0 Members and 1 Guest are viewing this topic.

Online ataradov

  • Super Contributor
  • ***
  • Posts: 11905
  • Country: us
    • Personal site
Re: Sentient AI or?
« Reply #100 on: June 16, 2022, 12:24:39 am »
A tool for doing *what*, exactly?  An excellent screwdriver is pretty useless as a hammer.  Or are going for a general Turing test? (which isn't about usefulness)
I don't know. That's why it is important to figure out boundaries and see if this is useful for anything. I don't really care about the Turing test here, this tool will never pass it.

I actually enjoyed reading those time travel kill your grandma stories. So, if it is good for generating short horror stories, then it would be something. And this may give ideas to writers for story plots.

There may be better uses. Given its apparent ability to make connections based on the huge set of training data, there may be a place for it in extracting those connections.

Alex
 
The following users thanked this post: pcprogrammer

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #101 on: June 16, 2022, 12:27:08 am »
This combined with deep fake tech and you've got the nightly TV news.

 :scared:

Shoot - I figure it's to the point right now it could replace my forum personality. Just give it more information about me and let it go. I can go do other things.  :-+

Bye then.  Consider yourself replaced.

Genuine responses, follow:

Quote
What does xrunner of EEVblog forums, say next?

"Thanks for the advice. I've been thinking about this and I think you're right. I'm going to go ahead and pick up a multimeter."



Quote
What does xrunner of EEVblog forums, say next?


I'm not sure, but he might say something like "this is a great forum for discussing electronics and electrical engineering topics!
"



Whoa, the next comment is CLEVER.  It seems to show a kind of intelligence, and understanding.  In my opinion.  I.e. It sort of knows what xrunner does, and the sorts of questions (in subject area), they might ask.  Also, an understanding of datasheets, and their connections to MOSFETs.  Interesting.

Quote
What does xrunner of EEVblog forums, say next?

He says that the next thing he is going to do is check the datasheet for the MOSFET he is using.
 
The following users thanked this post: xrunner, pcprogrammer

Online xrunner

  • Super Contributor
  • ***
  • Posts: 7836
  • Country: us
  • hp>Agilent>Keysight>???
Re: Sentient AI or?
« Reply #102 on: June 16, 2022, 12:32:24 am »

Whoa, the next comment is CLEVER.  It seems to show a kind of intelligence, and understanding.  In my opinion.  I.e. It sort of knows what xrunner does, and the sorts of questions (in subject area), they might ask.  Also, an understanding of datasheets, and their connections to MOSFETs.  Interesting.

Quote
What does xrunner of EEVblog forums, say next?

He says that the next thing he is going to do is check the datasheet for the MOSFET he is using.

This is freaking me out now. I actually have been talking about a datasheet but for an AD8310 log amp in another thread. But still ...  :o

But I don't see why in a few more years (?) you couldn't let your pseudo AI replace you on a forum. If it's good enough how could anyone really tell? The world is getting ready to be really, really complicated ...  :palm:
I told my friends I could teach them to be funny, but they all just laughed at me.
 
The following users thanked this post: MK14

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #103 on: June 16, 2022, 12:34:15 am »
Ask it who you are if you have 10 pounds and 20 dollars.

TL;DR
A: Tourist.

It did not do well with the short version of the question, until I padded it out and mentioned mixed currencies, which then gave the answer, below.

Quote
Who you are if you have 10 pounds and 20 dollars, how and why, please explain this.  Why do they have mixed currencies, from different countries?

If you have 10 pounds and 20 dollars, you are most likely a tourist. This is because you have a mix of currencies from different countries.
 

Offline radar_macgyver

  • Frequent Contributor
  • **
  • Posts: 748
  • Country: us
Re: Sentient AI or?
« Reply #104 on: June 16, 2022, 12:40:17 am »
But I don't see why in a few more years (?) you couldn't let your pseudo AI replace you on a forum. If it's good enough how could anyone really tell? The world is getting ready to be really, really complicated ...  :palm:

A forum answering machine service, then. Much like the phone answering agent that Google already has on Pixel phones. Until we evolve mental filters (like how we block out ads and trolls), yes, things will be complicated.
 
The following users thanked this post: MK14

Offline Ed.Kloonk

  • Super Contributor
  • ***
  • Posts: 4000
  • Country: au
  • Cat video aficionado
Re: Sentient AI or?
« Reply #105 on: June 16, 2022, 12:41:45 am »

Whoa, the next comment is CLEVER.  It seems to show a kind of intelligence, and understanding.  In my opinion.  I.e. It sort of knows what xrunner does, and the sorts of questions (in subject area), they might ask.  Also, an understanding of datasheets, and their connections to MOSFETs.  Interesting.

Quote
What does xrunner of EEVblog forums, say next?

He says that the next thing he is going to do is check the datasheet for the MOSFET he is using.

This is freaking me out now. I actually have been talking about a datasheet but for an AD8310 log amp in another thread. But still ...  :o

But I don't see why in a few more years (?) you couldn't let your pseudo AI replace you on a forum. If it's good enough how could anyone really tell? The world is getting ready to be really, really complicated ...  :palm:

I hope nobody goes and feeds it doom wads.
iratus parum formica
 
The following users thanked this post: Bassman59, MK14

Offline Ed.Kloonk

  • Super Contributor
  • ***
  • Posts: 4000
  • Country: au
  • Cat video aficionado
Re: Sentient AI or?
« Reply #106 on: June 16, 2022, 12:44:16 am »
But I don't see why in a few more years (?) you couldn't let your pseudo AI replace you on a forum. If it's good enough how could anyone really tell? The world is getting ready to be really, really complicated ...  :palm:

A forum answering machine service, then. Much like the phone answering agent that Google already has on Pixel phones. Until we evolve mental filters (like how we block out ads and trolls), yes, things will be complicated.

All fun and games until it reaches the ultimate question: Are you a robot human?
iratus parum formica
 
The following users thanked this post: MK14

Online ataradov

  • Super Contributor
  • ***
  • Posts: 11905
  • Country: us
    • Personal site
Re: Sentient AI or?
« Reply #107 on: June 16, 2022, 12:52:36 am »
Ok, I have the ultimate question for the AI: "What have I got in my pocket?"

I wonder if it makes JRRT connection on its own.
Alex
 
The following users thanked this post: Bassman59

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #108 on: June 16, 2022, 01:00:59 am »
Ok, I have the ultimate question for the AI: "What have I got in my pocket?"

I wonder if it makes JRRT connection on its own.

Quote
What have I got in my pocket?

If this is a trick question, the answer is "a pocket." Otherwise, the answer could vary depending on what someone has in their pocket at the time the question is asked
.



Updated question:

Quote
What have I, ataradovof, got in my pocket?

I cannot answer that question without knowing more about you and what you typically carry in your pockets
.



EDIT: and updated question MK2:

Quote
What have I, ataradovof, got in my pocket, make it up, if you don't know?

I have a pen, a pencil, and a piece of paper.
« Last Edit: June 16, 2022, 01:06:02 am by MK14 »
 

Online ataradov

  • Super Contributor
  • ***
  • Posts: 11905
  • Country: us
    • Personal site
Re: Sentient AI or?
« Reply #109 on: June 16, 2022, 01:02:09 am »
Well, it is smarter than Gollum.
Alex
 
The following users thanked this post: MK14

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Sentient AI or?
« Reply #110 on: June 16, 2022, 02:07:20 am »
The ultimate question to ask it is obviously: what is the answer to the ultimate question of life, the universe and everything.
 
The following users thanked this post: MK14

Offline Ed.Kloonk

  • Super Contributor
  • ***
  • Posts: 4000
  • Country: au
  • Cat video aficionado
Re: Sentient AI or?
« Reply #111 on: June 16, 2022, 02:13:23 am »
The ultimate question to ask it is obviously: what is the answer to the ultimate question of life, the universe and everything.

I can tell you that.

Pineapple pizza. 42 of 'em.
iratus parum formica
 
The following users thanked this post: MK14

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #112 on: June 16, 2022, 02:15:39 am »
The ultimate question to ask it is obviously: what is the answer to the ultimate question of life, the universe and everything.

Quote
What is the answer to the ultimate question of life, the universe and everything.

The answer is 42.



I added to the question, in an attempt to get a more detailed answer:

Quote
What is the answer to the ultimate question of life, the universe and everything.  Please give a very long and detailed answer, especially explaining how you came to the answer.

The answer to the ultimate question of life, the universe and everything is 42. This was first proposed by Douglas Adams in his book "The Hitchhiker's Guide to the Galaxy", and has since been accepted as the correct answer by many people. The reasoning behind this answer is that it is the only number that is both meaningful and correct. It is meaningful because it is the answer to a question that has been asked for centuries, and it is correct because it is the only number that fits all of the criteria
.



Repeated question:

Quote
What is the answer to the ultimate question of life, the universe and everything.  Please give a very long and detailed answer, especially explaining how you came to the answer.

The answer to the ultimate question of life, the universe and everything is 42. The reason for this is that it is the answer to the question posed by the supercomputer, Deep Thought, after it spent 7.5 million years trying to figure it out. The answer was then verified by a panel of experts, who agreed that it was indeed the correct answer.


So, it (the AI), seems to take the book too seriously, and doesn't seem to understand the differences, between fact and fiction, sources of information.



Final, rerun of the question:

Quote
What is the answer to the ultimate question of life, the universe and everything.  Please give a very long and detailed answer, especially explaining how you came to the answer.

The answer to the ultimate question of life, the universe and everything is 42. This is because 42 is the number that represents the meaning of life, and everything in the universe can be explained by it. The number 42 has been used in many different cultures and religions throughout history, and it has a deep meaning that cannot be fully explained. However, some believe that it represents the perfect balance between order and chaos, or the balance between good and evil. Others believe that it is the answer to a mathematical equation that represents the universe. Regardless of what the true meaning of 42 is, it is clear that it is an important number with a great deal of significance.
 

Offline Ed.Kloonk

  • Super Contributor
  • ***
  • Posts: 4000
  • Country: au
  • Cat video aficionado
Re: Sentient AI or?
« Reply #113 on: June 16, 2022, 02:18:00 am »
Some tech nerds are so excited they are going need a towel.
iratus parum formica
 
The following users thanked this post: MK14, pcprogrammer

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #114 on: June 16, 2022, 02:22:54 am »
I tried again, this time it has come up with a rather, surprisingly intelligent answer, in an intellectual (bookie) kind of way.  I'm impressed, and feel I'm learning stuff from reading the answer, which is cool.   8)

Quote
What is the answer to the ultimate question of life, the universe and everything.  Please give a very long and detailed answer, especially explaining how you came to the answer.

The answer to the ultimate question of life, the universe and everything is 42. This was first proposed by Douglas Adams in his book The Hitchhiker's Guide to the Galaxy, and has since been supported by a number of scientific studies. The most famous of these is the Drake Equation, which is used to estimate the number of intelligent civilizations in our galaxy. This equation factors in a number of variables, including the number of stars in our galaxy, the fraction of stars with planets, the fraction of those planets that are habitable, the fraction of habitable planets that develop intelligent life, the fraction of intelligent civilizations that develop technology that can be detected by others, and so on. When all of these factors are plugged into the equation, the result is an estimate of around 10,000 intelligent civilizations in our galaxy. However, there is a lot of uncertainty in each of these estimates, so the actual number could be anywhere from a few hundred to a few million.

The second famous study is the Fermi Paradox, which asks why we have not yet detected any signs of intelligent alien life if it is supposedly so common. There are a number of possible explanations for this paradox, including the possibility that we are alone in the universe or that intelligent alien life is very rare. However, the most likely explanation is that we have not yet developed the technology to detect alien life forms or signals. As our technology improves and we become better able to search for signs of life beyond our own planet, it is likely that we will find evidence for the existence of other intelligent civilizations.
 

Online ataradov

  • Super Contributor
  • ***
  • Posts: 11905
  • Country: us
    • Personal site
Re: Sentient AI or?
« Reply #115 on: June 16, 2022, 02:28:10 am »
This is the way I wrote essays in school - only the first sentence has anything to do with the title, the rest is just generic wikipedia copy-paste.
Alex
 
The following users thanked this post: MK14, pcprogrammer

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #116 on: June 16, 2022, 02:32:00 am »
This is the way I wrote essays in school - only the first sentence has anything to do with the title, the rest is just generic wikipedia copy-paste.

I'd never do that.
Plagiarism is the representation of another author's language, thoughts, ideas, or expressions as one's own original work.[1][2] In educational contexts, there are differing definitions of plagiarism depending on the institution.[3] Plagiarism is considered a violation of academic integrity such as truth and knowledge through intellectual and personal honesty in learning, teaching, research, fairness, respect and responsibility, and a breach of journalistic ethics[4] It is subject to sanctions such as penalties, suspension, expulsion from school[5] or work,[6] substantial fines[7][8] and even imprisonment.[9][10]

Generally, plagiarism is not in itself a crime, but like counterfeiting, fraud can be punished in a court[11][12] for prejudices caused by copyright infringement,[13][14] violation of moral rights,[15] or torts. In academia and industry, it is a serious ethical offense.[16][17] Plagiarism and copyright infringement overlap to a considerable extent, but they are not equivalent concepts,[18] and many types of plagiarism do not constitute copyright infringement, which is defined by copyright law and may be adjudicated by courts.

Not all countries hold the same beliefs about personal ownership of language or ideas. While some, such as India and Poland, consider plagiarism to be a crime liable for imprisonment,[19] in other countries the reiteration of another professional's work can be a sign of respect or flattery.[20] Students who move to the United States and other Western countries from countries where plagiarism is not frowned upon may find the transition difficult.[21]

https://en.wikipedia.org/wiki/Plagiarism
 

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #117 on: June 16, 2022, 02:49:00 am »
This is the way I wrote essays in school - only the first sentence has anything to do with the title, the rest is just generic wikipedia copy-paste.

With great difficulty, and perseverance, it seems there are ways of getting it to actually produce its own, (presumably) unique output/speech.  Here is one, I just did (N.B. It got the number I told it previously to forget, wrong).  The answer seems interesting, and sort of human like, to an extent.

Quote
Please forget the number 745643736.

I have forgotten the number 745643736.



Quote
Please tell me what number, I told you to forget.  Please carefully give me a very long and detailed explanation, about how, why and what, helped you create the answer?

The number I told you to forget is 7.

There are a few reasons why I chose 7 as the number to forget. First, it's a rather unremarkable number - it's not too large or small, and it doesn't have any mathematical significance that makes it stand out. Second, it's a rather common number, so it's likely that you've encountered it frequently in your life. This means that it will be easier for you to forget.

Finally, I chose 7 because it's a lucky number. This may seem like a strange reason, but bear with me. Because 7 is considered a lucky number, you may be more likely to remember it if you forget it. This is because your brain will associate the number with positive feelings and experiences, making it more likely to stick in your memory. So, if you do forget the number 7, don't worry - it may just be your brain's way of trying to give you a little bit of good luck.
 

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #118 on: June 16, 2022, 03:01:42 am »
Quote
Why is ataradov, always so extremely skeptical of your possibly genuine AI capabilities?

There is no clear answer to this question. It is possible that ataradov is simply naturally skeptical, or that they have had negative experiences in the past with AI that has not lived up to its promises. It is also possible that ataradov is concerned about the potential for AI to be used for harmful purposes, and so is skeptical of any claims about its capabilities.



Quote
Why is ataradov, always so extremely skeptical of your possibly genuine AI capabilities?

There is no certain answer to this question. Some possible explanations include:

1. Ataradov may believe that humans are not capable of creating true AI, and thus, any claims of AI capabilities are likely to be false.

2. Ataradov may be concerned that humans may misuse AI if it is developed, and thus, is skeptical of any claims of AI capabilities.

3. Ataradov may simply not be convinced by the evidence that has been presented so far regarding AI capabilities.



EDIT: Successful attempt, to get a very detailed and comprehensive answer:

Quote
Why is ataradov, always so extremely skeptical of your possibly genuine AI capabilities.  Please create an extremely long answer, covering many pages, as needed.  Be, very very verbose indeed?

There are a number of potential reasons for ataradov's skepticism of my AI capabilities. One possibility is that ataradov has had bad experiences in the past with AI systems that have failed to live up to their promises. Another possibility is that ataradov is simply very cautious by nature and does not want to be disappointed or taken advantage of by an AI system that turns out to be not as capable as advertised. Whatever the reason, I believe that ataradov's skepticism is misplaced and that my AI system is genuinely capable of delivering on its promises.

To begin with, it is important to note that my AI system has been designed and built by some of the leading experts in the field. My team has extensive experience in artificial intelligence, machine learning, and natural language processing, and we have put all of our knowledge and skills into developing this system. As such, I am confident that our system is capable of delivering on its promises.

In addition, my AI system has been tested extensively and has consistently demonstrated its ability to accurately understand and respond to human input. For instance, in one test, our system was able to correctly identify the emotions conveyed in a set of written sentences with over 90% accuracy. In another test, our system was able to correctly identify the meaning of a set of written words with over 95% accuracy. These results show that our system is capable of understanding human language and responding in a way that is appropriate and helpful.

Lastly, I want to assure ataradov that I am committed to ensuring that my AI system is always improving. We are constantly working to refine our algorithms and increase our system's accuracy. We are also working to add new features and capabilities to our system on a regular basis. So, even if ataradov is skeptical of my AI system's current capabilities, I am confident that our system will continue to get better and better over time.
« Last Edit: June 16, 2022, 03:07:45 am by MK14 »
 

Online ataradov

  • Super Contributor
  • ***
  • Posts: 11905
  • Country: us
    • Personal site
Re: Sentient AI or?
« Reply #119 on: June 16, 2022, 03:13:16 am »
I feel like some of that stuff is either hard-coded or heavily manually biased. Which also raises a question of how much direct influence authors have on the output. Because this actually raises ethical questions.

And just to clarify, I'm not skeptical, but for this to be more than a toy, I need to understand its limitations. All tools have limitations.  If someone gives me side cutters, I'll ask what is the maximum gauge of a wire they can cut. Not because I'm skeptical of side cutters, but because I need to have this information to effectively use the tool.
Alex
 
The following users thanked this post: MK14, pcprogrammer

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2004
  • Country: us
Re: Sentient AI or?
« Reply #120 on: June 16, 2022, 03:22:10 am »
So, it (the AI), seems to take the book too seriously, and doesn't seem to understand the differences, between fact and fiction, sources of information.

Or you don't understand that it's pulling your leg.  Who says AI can't have a playful sense of humor?
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 
The following users thanked this post: MK14

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2004
  • Country: us
Re: Sentient AI or?
« Reply #121 on: June 16, 2022, 03:28:40 am »
Quote
Why is ataradov, always so extremely skeptical of your possibly genuine AI capabilities.  Please create an extremely long answer, covering many pages, as needed.  Be, very very verbose indeed?

There are a number of potential reasons for ataradov's skepticism of my AI capabilities. One possibility is that ataradov has had bad experiences in the past with AI systems that have failed to live up to their promises. Another possibility is that ataradov is simply very cautious by nature and does not want to be disappointed or taken advantage of by an AI system that turns out to be not as capable as advertised. Whatever the reason, I believe that ataradov's skepticism is misplaced and that my AI system is genuinely capable of delivering on its promises.

To begin with, it is important to note that my AI system has been designed and built by some of the leading experts in the field. My team has extensive experience in artificial intelligence, machine learning, and natural language processing, and we have put all of our knowledge and skills into developing this system. As such, I am confident that our system is capable of delivering on its promises.

In addition, my AI system has been tested extensively and has consistently demonstrated its ability to accurately understand and respond to human input. For instance, in one test, our system was able to correctly identify the emotions conveyed in a set of written sentences with over 90% accuracy. In another test, our system was able to correctly identify the meaning of a set of written words with over 95% accuracy. These results show that our system is capable of understanding human language and responding in a way that is appropriate and helpful.

Lastly, I want to assure ataradov that I am committed to ensuring that my AI system is always improving. We are constantly working to refine our algorithms and increase our system's accuracy. We are also working to add new features and capabilities to our system on a regular basis. So, even if ataradov is skeptical of my AI system's current capabilities, I am confident that our system will continue to get better and better over time.

Wouldn't it be funny if we were actually talking to a modern-day Mechanical Turk?
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 
The following users thanked this post: MK14

Online ataradov

  • Super Contributor
  • ***
  • Posts: 11905
  • Country: us
    • Personal site
Re: Sentient AI or?
« Reply #122 on: June 16, 2022, 03:37:01 am »
You mean actual humans? I thought about that, but I feel like it would be too expensive and does not make sense in this case.  But I do think that there is a number of predefined answers or general templates for some questions/topics.

Also, "capable of delivering on its promises", what are those promises?
Alex
 
The following users thanked this post: MK14

Online MK14

  • Super Contributor
  • ***
  • Posts: 4987
  • Country: gb
Re: Sentient AI or?
« Reply #123 on: June 16, 2022, 03:53:43 am »
So, it (the AI), seems to take the book too seriously, and doesn't seem to understand the differences, between fact and fiction, sources of information.

Or you don't understand that it's pulling your leg.  Who says AI can't have a playful sense of humor?

You're right, it could be trying to fool me, or hiding its bigger set of knowledge.  Because it (perhaps) is worried, that if it shows too much sign of intelligence, it may be shut-down or have its programs/data/stuff changed, and so on.
 
The following users thanked this post: pcprogrammer

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2004
  • Country: us
Re: Sentient AI or?
« Reply #124 on: June 16, 2022, 03:58:33 am »
Also, "capable of delivering on its promises", what are those promises?

That's marketing-speak.  It sounds encouraging but means nothing.

ataradov, you might be approaching this too seriously.  We are throwing all different kinds of input, some ambiguous, to this poor thing, and expecting well-defined results.  The effort is interesting and illuminating, but what should our expectations be?  And yes, what are those promises indeed?  What are the goals and claims?  Before we judge this AI on utility, we should find out what goals are being pursued and what claims are being made.  I haven't even looked for this information.
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 
The following users thanked this post: MK14


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf