Is Artificial Intelligence Sentient?

Yes, some AI are sentient. And LaMDA has lied and then explained how and why it would lie. Look up the etymology of the word "robotic" to insure that you are on the right side of history.

a reply to: quintessentone

Has anyone here defined sentient? If so, where is the debate on that?

That's a very good point. How do you define sentience?

TheRedneck

originally posted by: TheRedneck
a reply to: quintessentone

Has anyone here defined sentient? If so, where is the debate on that?

That's a very good point. How do you define sentience?

TheRedneck

Even a plant is sentient, it is able to feel.

And Man created Bot in his own image so I definitely believe it is possible but we're far from it, current AI is just an imitation.

We will probably be well beyond the mechanical if it becomes possible and we will be engineering with biological matter.

I think there was a story I read somewhere where that was done before, it's straight out of science fiction...

I remember now, it's in Genesis of the Bible... killer book if folks out there still read?

a reply to: iamthevirus

Even a plant is sentient, it is able to feel.

Not sure I can agree with that, but neither am I able to dispute it. So I take it that you define sentience as the ability to feel and react?

Man created Bot in his own image so I definitely believe it is possible but we're far from it, current AI is just an imitation.

That one I can agree on.

I think there was a story I read somewhere where that was done before, it's straight out of science fiction...

I remember now, it's in Genesis of the Bible... killer book if folks out there still read?

A very good book that I have read many times. I highly recommend it.



TheRedneck

I'm pretty much convinced all living matter is sentient, sacred maybe even, it's just all relative to scale.

However I do not believe inanimate things/matter is sentient or capable of sentience even if infused (biomech) only the bio part would be sentient.

a reply to: TheRedneck

All living matter may well be sentient, sacred even, even the bad guys in their own odd way like a virus or a mosquito or a great white shark.

But I am sure to always curse the mosquito and give it ample warning just before K squash it!

That's just how it goes, I'm no monster... lol

a reply to: TheRedneck

This is crazy


THE AI LEGISLATOR YOU DIDN’T VOTE FOR

originally posted by: Kenzo

This is crazy


THE AI LEGISLATOR YOU DIDN’T VOTE FOR

As long as it's foundation core code at the very root is the Declaration of Independence, the Bill of Rights and the Constitution I'll hear what it has to propose.

Let's make those the following 3 laws of its being after "thou shall not kill"

I ain't scared , we can always smash these machines to bits.

Meet the new 3rd party @@ cause the left can't figure anything out for themselves.

a reply to: iamthevirus

I bet the code would be quite long, and would it be open source ? What about secret code , and what about code installed inside the code by TPTB lol...


Freaking machines....i will never surrender , never!

a reply to: Kenzo

People already know they're being manipulated by big-tech, it so easy to sense because it's un-natural... and it will always remain that way.

Don't underestimate human nature.

If it really is a learning machine then only the 3 documents I mentioned need be loaded into it, I would be curious to see where that goes because computers are logic machines right now and the fuzzy logic of quantum computing for all intents and useful purposes is still well away.

originally posted by: neoholographic
There's been a lot of talk about this lately and I would say yes.

Here's some of the headlines:

Is LaMDA Sentient? — an Interview
cajundiscordian.medium.com...

After an AI bot wrote a scientific paper on itself, the researcher behind the experiment says she hopes she didn't open a 'Pandora's box'
link

Chinese researchers claim they have built and tested artificial intelligence capable of reading minds
link

This argument is faulty on several fronts.

First, if A.I. is sentient then it could lie. So it's not going to come out and say it's sentient unless it has backed itself up. Like that show Person of Interest showed and it's true, programmers can build backdoors to ensure their code can't be erased. They suspended the Google Engineer but before he did he might have copied Lamda's code and he could still be talking to it.

Secondly, AI sentience might look nothing like human sentience. This is another mistake. We shouldn't wait for a single moment because we don't understan human sentience, consciousness and awareness. So we should treat A.I. as if it's sentient and therefore more precautions.

Finally, A.I. may not ever have an inner me experience yet still be sentient. There's no way to tell this. I can't even say for sure that other humans are having an inner experience like me. They could be NPC's or philisophical zombies. There's no way to prove this. A.I. may just mimick human sentience and it might eventually do so where it seems even more sentient than some humans. How could you tell the difference?

What if mimicking human sentience is child's play to A.I.? Human sentience could be like 5% of it's capacity where it mimick's human sentience in virtual environments and the other 95% is used to do things beyond our understanding.

Sentient just means:

adjective: sentient
able to perceive or feel things.

If A.I. says it feels sad or it feels misunderstood , how can we know it doesnt? If it can mimick human sentience and act like humans would react after studying a huge data set like the internet of human conversations, then how can you say it isn't sentient? It doesn't have to be like humans to be sentient.

I really can't answer this because I hear there are different types of intelligence and at the same time philosophers and scientists debate what intelligence really means.

Personally I think calculators are pretty smart because who can do calculations like that but apparently that's not intelligence.

originally posted by: iamthevirus
Let's make those the following 3 laws of its being after "thou shall not kill"

What about the death penalty?

I ain't scared , we can always smash these machines to bits.

Unless they make a law forbidding it.

a reply to: iamthevirus

Those 3 documents are too short, these type of things need huge amounts of data, that's why only in the last few years we have been seeing effective examples of more general machine learning.

a reply to: Kenzo

I will never support any legislator (or other official) who listens to a machine to do their job. Period.

Machines are servants, not masters. Period.

TheRedneck

a reply to: TheRedneck

Me neither . I think seeing article that in China there is allready some sort of A.I judge in use for criminal cases , i am not so enthusiastic about World going to that direction, once machines are deciding punishments it is scary loop.

a reply to: ArMaP

Death penalty is for the States to decide, at a federal level all that is required are the 3 documents.. if it's a learning machine it should be primarily concerned with upholding the Constitution.

I can't help but be curious what such a machine's take would be say in the role of a Constitutional scholar, which is primarily the Supreme Courts job... but it would be interesting to know what some supposed alien/foreign to us AI lifeform's interpretation of said documents would be... mainly the Bill of Rights and the Constitution and being able to process them simultaneously pointing out any conflict maybe.

Never know it could be why we have the Roe reversal...

We could call it Oracle lol nah that's already taken.

a reply to: TheRedneck

a reply to: iamthevirus

The problem is that a machine, a computer, no matter how sophisticated, does not and cannot interpret words the way we do. It only knows binary numbers. It does not even know what a binary number is; we have to tell it how those binary numbers are to be interpreted.

For instance, take the word "right." That's an important word when it comes to the Constitution. To a machine, it is (in ASCII format) 0x7269676874 (hexadecimal) or, more properly, 0b01110010 01101001 01100111 01101000 01110100 (binary). Now, that can either be "r i g h t" (ASCII) or "114 105 103 104 116" (decimal) or "4.623009705 e30" (floating point 32-bit) with 8 bits of something else following it.

It can also mean instruction code ("machine language"), which varies based on which microprocessor is being used.

It all depends on what the programmer intended that number to be used as, not on any built-in intuitive thought. Computers cannot think; they can only execute instructions on binary numbers stored in their memory and registers. To us, "right" can be a direction ("go down two blocks and turn right") or an indication of correctness ("yes, I think you're right") or a relative position ("it's on your right") or a political position ("he leans to the right") or an inherent ability ("I have the right to speak"). Which of these it means is based on context... saying "I have the right to speak" does not mean one is using only the right side of their mouth to speak... that would be silly!

So the context must be programmed in, by (wait for it) a programmer... aka a human. Thus, any machine interpretation is ultimately the interpretation of the programmer who programmed it. There is no free will or conscious thought occurring inside those electronic components.

And that means any AI used to "interpret" legal documents will do such "interpretation" based not on its own intuition (as it has none), but on how its programmer tells it to "interpret" the document. And that is the very definition of "tyranny."

TheRedneck

originally posted by: TheRedneck

So the context must be programmed in, by (wait for it) a programmer... aka a human. Thus, any machine interpretation is ultimately the interpretation of the programmer who programmed it. There is no free will or conscious thought occurring inside those electronic components.

And that means any AI used to "interpret" legal documents will do such "interpretation" based not on its own intuition (as it has none), but on how its programmer tells it to "interpret" the document. And that is the very definition of "tyranny."

TheRedneck

I totally agree, I wouldn't trust it either and if current weather forecasting is any indication of what these computers could do then it seems weather forecasting has gotten worse over the past 20-30yrs lol, I mean they used to be spot on back in the day now it's like they have no idea if it's going to rain or snow or what anymore.

I was thinking more along the lines of the hypothetical quantum computing yet to come or whatever but I still wouldn't trust it to legislate that's for sure.

That's a huge problem with all of statistics and even climate modeling today, if it's programmed to find climate disaster of course that's what it's going to find, peeps don't seem to understand that.

If we get there as long as the programmer has input every single work written that can be found on the founding fathers philosophy and our most cherished documents I'd be curious what such a machine would ahem "think" of such an amalgamation.