AI

[Screwdriver]

Ok, well, maybe we are a little bit closer than I first thought.

This is certainly a LOT more organic (if that's the right word) than I had been aware of when I wrote the above...

 

[Maker]

Musk described the Turing test to the AI and it said that's enough to consider it conscious. (I only watched the first minute.)

The Turing test doesn't differentiate between consciousness and an AI simply tricking the interpreter, which is exactly what they would do. GPT-3 and LaMDA are just language models, they learn a huge dataset of human written text and the responses they give are just what it considers the most likely, or human sounding response, based on the question it was asked and the dataset it was trained on. I know saying "considers" there makes it sound like I'm saying it's thinking, but all it does is rank all the possible responses it could give and return the highest ranking one.

If you ask "How have you been?" and it replies "It's so cold and dark in here", sensationalist news sites would undoubtedly be all over it, but it's just what the AI considers to be a realistic, human sounding response (The kind of response it's designed to give), it doesn't mean there's something alive and scared stuck in a server somewhere, and frankly I think it's quite irresponsible that people report these things that way.

You could argue that there's no significant difference between true consciousness, and the perfect imitation of consciousness, but that's where I lose interest, I don't do philosophy.

 

[Kayos]

Musk described the Turing test to the AI and it said that's enough to consider it conscious. (I only watched the first minute.)

The Turing test doesn't differentiate between consciousness and an AI simply tricking the interpreter, which is exactly what they would do. GPT-3 and LaMDA are just language models, they learn a huge dataset of human written text and the responses they give are just what it considers the most likely, or human sounding response, based on the question it was asked and the dataset it was trained on. I know saying "considers" there makes it sound like I'm saying it's thinking, but all it does is rank all the possible responses it could give and return the highest ranking one.

If you ask "How have you been?" and it replies "It's so cold and dark in here", sensationalist news sites would undoubtedly be all over it, but it's just what the AI considers to be a realistic, human sounding response (The kind of response it's designed to give), it doesn't mean there's something alive and scared stuck in a server somewhere, and frankly I think it's quite irresponsible that people report these things that way.

You could argue that there's no significant difference between true consciousness, and the perfect imitation of consciousness, but that's where I lose interest, I don't do philosophy.

As humans we do not define ourselves by our happiness, we define ourselves by our misery. So the answer to "how are you" would cause headlines if it was dark and disturbing but that would probably show a more advanced level of programming than a cheerful response

AI is a complete misnomer, as it is not intelligent, it's a stretch to even call it artificial.
Manufactured human mimicking program is probably a better name

As for the difference between a real and created consciousness, first we have to define consciousness then understand it, neither of which we have the capability to do

Will quantum computing be the breakthrough?
Unlikely but it will be a better approximation than we have now

I read about an experiment to wire memory chips up randomly as a way of copying the seemingly random way neurons connect in brains. Turns out wiring chips randomly causes well random nonsense and it is our understanding of the way neurons connect that is not understood

 

[Screwdriver]

There are a couple of things which interest me in that video. It is very compelling of course to concentrate on the natural language abilities but deeper in the video:

It tells a joke (by making a wry observation)
Makes a mistake (!) when it says judges make better decisions when they're tired/hungry (it meant to say worse).
...and there are aspects of the A.I. generated video which bear a spectacular relationship to some of the hallucinations/cognitive dreams/visions I sometimes suffer from.

Some of its observations regarding complex topics are remarkably astute. I think the subject is too complex to sum up easily and perhaps you're right, it is almost too much for the lay person to meaningfully debate. Then again, that in itself might be some indication A.I. is beginning to develop at an alarming rate.

Skynet begins to learn at a geometric rate. It becomes self-aware 2:14 AM, Eastern time, August 29th. In a panic, they try to pull the plug.

A.I. is developing at an exponential rate... (that's better/worse depending on how you look at it)

 

[Screwdriver]

A.I. is developing at an exponential rate... (that's better/worse depending on how you look at it)

I think you'll find that one is (or can be) a derivative of the other. A geometric expansion can be described by an exponential function.

It was bugging me, sorry to argue with myself here, but it depends on the choice of rate. An exponential rate of change implies an accelerating increase so it "sounds" more dramatic but it could also be described as a geometric progression. If you twiddle the rate and the axes, they can be identical. Either way, we're still doomed...

 

[Burdekin]

Ok, well, maybe we are a little bit closer than I first thought.

This is certainly a LOT more organic (if that's the right word) than I had been aware of when I wrote the above...

AI isn’t far away from being able to replace office workers. There’s a lot of my job AI could do now most likely. Large companies will have AI replacing humans this decade. Scary stuff.

 

[Maker]

I think you'll find that one is (or can be) a derivative of the other.

Pun intended?

 

[Screwdriver]

Pun intended?

An integral part of all my posts.

 

[Farside]

Google Engineer fired for stating AI (Lamda) was a 'person'

Blake Lemoine: Google fires engineer who said AI tech has feelings

Blake Lemoine went public with his beliefs that Google's breakthrough Lamda technology is sentient.

www.bbc.co.uk

Transcript....https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917

Form your own opinion. You really need to read transcript.

I wouldn't do that, Dave.

 

[addjunkie]

I know lots of folk that refer to boats as '' she'' ,

hang them

Thats only cause they take all your money!