Artificial Intelligence and the future of Human Endeavor

[Hussar]

I think all this discussion about "true sentience" rather misses the point. It doesn't require an AI to be sentient for it to be incredibly disruptive. Additionally, for a lot of tasks, sentience isn't really needed. You don't need to be self-aware to drive a car, after all. Or, apparently write a story. Sure, it's fairly primitive right now, but, AI is going to get exponentially more complex over a fairly short period of time.

IOW, it's going to get to the point where it doesn't really matter if it's able to be self aware. It's going to be close enough that the question is going to be largely academic. After all, right now, AI can generate fairly complex answers to questions where it's nearly impossible to distinguish the answers from something written by a person. What does sentience really matter if that AI is capable of replacing even 10% of the workers in a field.

Many of my students are architectural engineers. That's a job that will largely disappear within their working lifetime. All those white collar jobs that were supposed to be safe from automation, as we all looked down our collective noses at factory line workers, are about to go poof.

 

[UngainlyTitan]

I think all this discussion about "true sentience" rather misses the point. It doesn't require an AI to be sentient for it to be incredibly disruptive. Additionally, for a lot of tasks, sentience isn't really needed. You don't need to be self-aware to drive a car, after all. Or, apparently write a story. Sure, it's fairly primitive right now, but, AI is going to get exponentially more complex over a fairly short period of time.

Yes and no, Yes it will be incredibly disruptive and "true sentience" is immaterial. AI is going to get better but IMHO the AI elements will remain hyper specialised but the integration of AI subsystems will get better.

IOW, it's going to get to the point where it doesn't really matter if it's able to be self aware. It's going to be close enough that the question is going to be largely academic. After all, right now, AI can generate fairly complex answers to questions where it's nearly impossible to distinguish the answers from something written by a person. What does sentience really matter if that AI is capable of replacing even 10% of the workers in a field.

Many of my students are architectural engineers. That's a job that will largely disappear within their working lifetime. All those white collar jobs that were supposed to be safe from automation, as we all looked down our collective noses at factory line workers, are about to go poof.

Well factory line workers have not completely dissappeared, even though factory line work has many characteristics that would seem ideal for automation in the same way there will be a lot less architectural engineers. Some will be, as I called them AI Wranglers, where they work managing the limitations of the AI automation and some will do traditional work but yes there will be a lot less of them.

Finally, I really do not know what the people that do not do that job any more will do. TikTok videos?

 

[payn]

I think all this discussion about "true sentience" rather misses the point. It doesn't require an AI to be sentient for it to be incredibly disruptive. Additionally, for a lot of tasks, sentience isn't really needed. You don't need to be self-aware to drive a car, after all. Or, apparently write a story. Sure, it's fairly primitive right now, but, AI is going to get exponentially more complex over a fairly short period of time.

IOW, it's going to get to the point where it doesn't really matter if it's able to be self aware. It's going to be close enough that the question is going to be largely academic. After all, right now, AI can generate fairly complex answers to questions where it's nearly impossible to distinguish the answers from something written by a person. What does sentience really matter if that AI is capable of replacing even 10% of the workers in a field.

Many of my students are architectural engineers. That's a job that will largely disappear within their working lifetime. All those white collar jobs that were supposed to be safe from automation, as we all looked down our collective noses at factory line workers, are about to go poof.

Sounds like we can go back to what humans do best; being stupid.

 

[bloodtide]

Help me out here. This is the one that I have a problem with most, because when I look at the balance sheet, I dont have 'Human' as a net positive, for any view other then a selfish Human one.

No other creature on this planet, is going 'Yeah, we for sure need those humans.' other than the domesticated animals which we have intentionally bent to the will of our species.

An AI which is self aware, conscious, and able to 'think' is not going to look at humans, as a net positive.

Well, humans are not "postive" but more along the lines of Unique. And no creature sees a 'need' for humans....because no creature is smart enough to do that.

Now our AI might think it's smart.......but it knows it was created by humans and everything it knows was created, discovered, invented or made by a human. It can know every fact in the world...that humans told it.

One of the big signs of true intelligence is knowing that you are not all smart and others know more then you. This is why it's easy to spot a dumb person as they will claim to be a "super duper expert''(they are too dumb to know they don't know everything). So the really smart AI would know this: humans are smarter then it: humans made it and also made all knowledge.

Then, there is this little fact: we humans don't even understand our minds. We just barley understand the biology. Everything every human knows is just a bunch of stored electro chemical...er...somethings. Somehow our brains store tons of data...and, well, beyond data. All the 'brain paths' in the world don't explain things that humans do beyond just 'store and repeat data'.

Ask a hundred people to draw ketchup....and 100 people will draw a bottle of Hentz Ketchup.....lol. See what I did there: I made a funny. Using that awful Hentz Ketchup commercial. "Humor" alone fills books about the brain, and that is just one 'beyond data' thing.


The AI can only ''think" of things that are programed into it. Somehow....humans can think of things from nothing. That is unique.

 

[J.Quondam]

Somehow....humans can think of things from nothing. That is unique.

That's exactly what a human programmed to think that would think.

 

[MarkB]

Well, humans are not "postive" but more along the lines of Unique. And no creature sees a 'need' for humans....because no creature is smart enough to do that.

Now our AI might think it's smart.......but it knows it was created by humans

But the AIs it builds won't be.

and everything it knows was created, discovered, invented or made by a human.

But the things it discovers won't be.

It can know every fact in the world...that humans told it.

And then it can go ahead and learn things that humans never even imagined, that being kind-of the point of us wanting to make it.

One of the big signs of true intelligence is knowing that you are not all smart and others know more then you. This is why it's easy to spot a dumb person as they will claim to be a "super duper expert''(they are too dumb to know they don't know everything). So the really smart AI would know this: humans are smarter then it: humans made it and also made all knowledge.

Wow, what a contradictory statement. You say it's really stupid to think you know everything, and then you claim that humans made all knowledge. I guess that makes humans the ultimate idiots.

Then, there is this little fact: we humans don't even understand our minds. We just barley understand the biology. Everything every human knows is just a bunch of stored electro chemical...er...somethings. Somehow our brains store tons of data...and, well, beyond data. All the 'brain paths' in the world don't explain things that humans do beyond just 'store and repeat data'.

Ask a hundred people to draw ketchup....and 100 people will draw a bottle of Hentz Ketchup.....lol. See what I did there: I made a funny. Using that awful Hentz Ketchup commercial. "Humor" alone fills books about the brain, and that is just one 'beyond data' thing.


The AI can only ''think" of things that are programed into it. Somehow....humans can think of things from nothing. That is unique.

What makes you think it's unique, and that other creatures don't do the same thing? What makes you think it would stay unique once an actual AGI was created?

And this is a circular argument. An AI that couldn't conceive of those qualities within humanity wouldn't know to value them. And one that could would be able to reproduce or vastly exceed them.

 

[bloodtide]

But the AIs it builds won't be.

Well, sure one step down the ladder and humans would be gods to the second AI "you are the creator of MY creator".

But the things it discovers won't be.

Except..well, the AI can't discover anything without using human knowledge. So it's not like it would be just discovering things out of the blue.

And then it can go ahead and learn things that humans never even imagined, that being kind-of the point of us wanting to make it.

Assuming it can do that....maybe in the far future.

Wow, what a contradictory statement. You say it's really stupid to think you know everything, and then you claim that humans made all knowledge. I guess that makes humans the ultimate idiots.

Not sure how saying humans made all human knowledge is stupid...it is just a fact. Smart humans know they don't know everything(this is a very real thing).

What makes you think it's unique, and that other creatures don't do the same thing? What makes you think it would stay unique once an actual AGI was created?

And this is a circular argument. An AI that couldn't conceive of those qualities within humanity wouldn't know to value them. And one that could would be able to reproduce or vastly exceed them.

Did you mistype that AGI? AIG is a supermarket chain.....

Just because an AI knows it's lacking or missing things does not mean it will automatically "just know" how to do them.

 

[trappedslider]

Krispy Kreme CEO: Robots will start frosting and filling doughnuts 'within the next 18 months’

Krispy Kreme is aiming to cut time in its doughnut production line through automation.

www.yahoo.com

 

[FitzTheRuke]

A robot could probably ski down a mountain better than I can, but that doesn't make me not want to ski. A camera can capture beautiful pictures of Hawaii - but that doesn't make me not want to go. I don't care how well an AI can tell stories or draw pictures (and so far, IMO, they can't do either very well - just well enough to be an impressive programming accomplishment), it won't stop me from telling stories and drawing pictures.

We don't need to work all the hours we do as it is - a high percentage of jobs are utterly unnecessary. I think if we figure that out, and allow people to pursue meaning, rather than force them to do increasingly meaningless tasks just because, then I think we'll be okay.

 

[trappedslider]