How can machines learn?

[Persuader]

This ua what make AI a existential threat.....their ability to learn.

Machine learning automates analytical model building. It uses methods from neural networks, statistics, operations research and physics to find hidden insights in data without being explicitly programmed where to look or what to conclude.



AI could harm the health of millions and pose an existential threat to humanity, doctors and public health experts have said as they called for a halt to the development of artificial general intelligence until it is regulated.

The final barrier AI will have to cross to enable their ability to threaten mankind would be the ability to reproduce.

US scientists who created the first living robots say the life forms, known as xenobots, can now reproduce – and in a way not seen in plants and animals.

Do Some Passages in the Book of Revelation Seem to Talk About AI?

Oxford mathematician John Lennox notes that in Revelation, the freedom to buy and sell is determined by a kind of mark, like China’s social credit score.

mindmatters.ai

https://www.cnn.com/2021/11/29/americas/xenobots-self-replicating-robots-scn/index.html#:~:text=and%20self-heal.-,Now,-the%20scientists%20that

 

[Leweman]

It's something to be considered about in the future. It's not nothing.

 

[ReinyDays]

Can machines anticipate the thoughts in humans? ... not over the chessboard they can't ...

 

[Anomalism]

We're threatened by anything we can't explain or control. We have no reason to believe the AI will want to kill us.

 

[Persuader]

Can machines anticipate the thoughts in humans? ... not over the chessboard they can't ...

A team of scientists from the University of Texas at Austin has developed an AI model that can read your thoughts. The noninvasive AI system known as semantic decoder lays emphasis on translating brain activity into a stream of texts according to the peer-reviewed study published in the journal Nature Neuroscience.May 3, 2023

 

[Rambunctious]

How can machines learn?​

Since they don't attend our public schools they learn just fine....

 

[BULLDOG]

 

[BlindBoo]

This ua what make AI a existential threat.....their ability to learn.

Machine learning automates analytical model building. It uses methods from neural networks, statistics, operations research and physics to find hidden insights in data without being explicitly programmed where to look or what to conclude.



AI could harm the health of millions and pose an existential threat to humanity, doctors and public health experts have said as they called for a halt to the development of artificial general intelligence until it is regulated.

The final barrier AI will have to cross to enable their ability to threaten mankind would be the ability to reproduce.

US scientists who created the first living robots say the life forms, known as xenobots, can now reproduce – and in a way not seen in plants and animals.

Do Some Passages in the Book of Revelation Seem to Talk About AI?

Oxford mathematician John Lennox notes that in Revelation, the freedom to buy and sell is determined by a kind of mark, like China’s social credit score.

mindmatters.ai

https://www.cnn.com/2021/11/29/americas/xenobots-self-replicating-robots-scn/index.html#:~:text=and%20self-heal.-,Now,-the%20scientists%20that

Reproducing Mutant Frog Claws?

It's the end of the world as we know it for sure.

 

[ReinyDays]

That's a long build up ... typical Woody Allen ... but always worth the punch-line ...

But good question: will AI be anti-semitic? ... I sure hope not ...

 

[whitehall]

It's not learning as such, it's about following a program. A machine can't learn to murder humans but when you program a missile to react to a certain word or button or situation it will kill friend or foe.

 

[Resnic]

Machines learn through what is called machine learning. It isn't the same as humans which learning is done biomechanically. A computer does it through parameters set in place by humans.

As we advance, advance our software and advance the hardware machines can process a greater and greater amount of parameters. Potentially to the point a computer can start to write its own parameters and alter them as needed.

But a machine can't ever truly "learn". They aren't real. Even if one tried to convince you it was alive so to speak, it's still working on highly advanced sets of parameters and is computing how to act like a person that's alive.

Can machines anticipate the thoughts in humans? ... not over the chessboard they can't ...

The first time a computer beat a world chess champion was in 1996. It's been done multiple times since then.

https://cybernews.com/editorial/rise-of-the-machines-when-a-computer-beats-a-chess-master/

 

[there4eyeM]

Attention must be paid to what level of decision making "AI" might have.
In many areas, some of us believe we already have "AI"; absent intelligence. Artificial would at least be something.
Running traffic lights comes to mind.

 

[ReinyDays]

The first time a computer beat a world chess champion was in 1996. It's been done multiple times since then.

https://cybernews.com/editorial/rise-of-the-machines-when-a-computer-beats-a-chess-master/

Tactical engines ... I've played a few ... chess computers crunch out moves, combinations, and then picks the move that gives the best probability to win ...

That's different from strategic play ... put the knight on bishop 3 because it is a good place for the knight ... good strategy, not tactics ... try the French Defense and don't castle, this tends to give computers problems ... then pawn storm the computers king's position ... you'll get your draw if you play the endgame perfectly ...

 

[there4eyeM]

How can we get humans to learn?

 

[Persuader]

How can we get humans to learn?

The real problem is kids learning the wrong things.

I am not even going to talk about our public schools.....that discussion would require more time than I am willing to allocate.

 

[Quasar44]

AI does not exist in 2023.. nor will it anytime in the decades to come

 

[Persuader]

AI does not exist in 2023.. nor will it anytime in the decades to come

In a technical sense you may be right about 2023....yet the concept and or reality of what is called Artificial intelligence is rapidly evolving.

Artificial intelligence (AI)

Artificial intelligence or AI simply means software used by computers to mimic aspects of human intelligence. For example, a program that recommends what you should read based on books you’ve bought or a robot vacuum that has a basic grasp of the world around it.

www.newscientist.com

The rise of machine consciousness: Studying consciousness with computational models

Efforts to create computational models of consciousness have accelerated over the last two decades, creating a field that has become known as artifici…

www.sciencedirect.com

 

[Persuader]

Tesla and SpaceX tech mogul Elon Musk has strong opinions about artificial intelligence (AI), suggesting that AI may “summon the demon” that will put humanity at risk or turn us all into the house pets of machines. He believes AI may be even more dangerous than nuclear weapons. To fight the perceived dangers, Musk is planning to unveil new computers that can be placed directly in human brains, what some call “neural lace.” The phrase was first conceived by science fiction

 

[Persuader]

Entering the singularity: Has AI reached the point of no return?​

Entering the singularity: Has AI reached the point of no return?

AI will be effective not only when machines can do what humans do (replication), but when they can do it better and without human supervision .

thehill.com

 

[petro]

Headline 2034.
First sentient AI commits suicide after discovering humans created it.