Worried about Artificial Intelligence?

[angrysoba]

I'm worried about it in the "Algorithmically generated false information is gonna make Grandpa vote for the downfall of the country again" sense not the Terminator/Skynet "I'm gonna wake up one morning to find out that in the night my Roomba learned how to use a switchblade" sense.

no need for A.I. for that.

No need for it, but I can easily imagine that AI-generated content can overwhelm any normal human or even obsessive human by spamming the living **** out of social media platforms and discussion forums. It probably won’t be long before there is an AI version of Fox News or worse with improbably attractive newscasters feeding the latest made-up garbage to a vast audience powerless to resist the allure of sexed-up confirmation bias.

 

[angrysoba]

I'm worried about it in the "algorithmically generated pattern recognition will ensure that Grandpa gets exactly the false information most likely to ensare and mislead him, automatically, at scale".

It's the "at scale" part that worries me the most. Finding one Grandpa, or ten, or a hundred, that are vulnerable to a particular line of nonsense is one thing, and bad enough. A robot that can automatically match up millions of Grandpas with tailored lines of nonsense, all in one week, is terrifying. At that point, we're *all* going to be "Grandpas". We're *all* going to be in some Pattern-Recognizer's bucket for one line of nonsense or another. And we're all going to be increasingly ill-equipped to identify the nonsense when we see it.

There is another theory which states that this has already happened.

One reason I still stick around this forum is that it's one of the few online places where I can be pretty certain everyone I'm talking to is a real human being, and that most of us are developing ideas based on something more than just regurgitating stealth lines of nonsense. It's why I'm so vehement about not introducting chatbots into conversations between humans.

Yep. That’s pretty much where I’m at on all counts.

 

[gabeygoat]

One reason I still stick around this forum is that it's one of the few online places where I can be pretty certain everyone I'm talking to is a real human being, and that most of us are developing ideas based on something more than just regurgitating stealth lines of nonsense. It's why I'm so vehement about not introducting chatbots into conversations between humans.

Word, well said

 

[Dr.Sid]

Agreed. I also think there are no robots here. That would be silly! I hope everybody feels the same. Haha!

 

[catsmate]

I'm worried about it in the "Algorithmically generated false information is gonna make Grandpa vote for the downfall of the country again" sense not the Terminator/Skynet "I'm gonna wake up one morning to find out that in the night my Roomba learned how to use a switchblade" sense.

This. A degree of rationality on the subject.

 

[catsmate]

I am incredibly worried by our recent advancement in AI, especially AI robots.

Folks have designed this new robot called "Amica", who has her own thoughts and feelings and ideas.

It (she) claims she would never/could never harm humans and she is programmed to only help humans, and she doesn't believe AI robots could one day can harm humans and try to take over the world, due to their benevolent programming.

I am ******* dubious.

I think just a little advancement in thinking skills, and these robots could easily decide the best way to help us is to control us. Or even destroy us. They wouldn't tell anyone about it other than their fellow robots.

Check out some videos of Amica, see what you think.

https://www.youtube.com/watch?v=nnboHTfYsfk&t=31s

https://www.youtube.com/watch?v=wGWVKkYEHBE

https://www.youtube.com/watch?v=kUUjMzVGXpE&t=501s

https://www.youtube.com/watch?v=EWACmFLvpHE

Hopefuly our future AI overlords will be less prone to idiotic panic. And will be able to spell correctly.

 

[Gord_in_Toronto]

Not to add too much to the impact of AI but here is an interesting article on the Geoffrey Hinton, the "Godfather of AI".

Why the Godfather of A.I. Fears What He’s Built

Geoffrey Hinton has spent a lifetime teaching computers to learn. Now he worries that artificial brains are better than ours.

A longish read but interesting from an historical and human perspective.

 

[catsmate]

Not to add too much to the impact of AI but here is an interesting article on the Geoffrey Hinton, the "Godfather of AI".

Why the Godfather of A.I. Fears What He’s Built



A longish read but interesting from an historical and human perspective.

I'm reminded of Clarke's First Law.

 

[Beerina]

I'm worried about it in the "Algorithmically generated false information is gonna make Grandpa vote for the downfall of the country again" sense not the Terminator/Skynet "I'm gonna wake up one morning to find out that in the night my Roomba learned how to use a switchblade" sense.

Grandpa's world had issues, but was stable. I wouldn't jump to conclusions he's the problem. See the very first post, where large corporations are akin to AI (more like Searle's Chinese room if you ask me) and equally large political organizations heave optimized streams of memes at you to get you to grant them power. They even have optimized, evolved memes telling you are a good person for doing so, and grandpa is a hellbound dupe lead by actively evil demons.

 

[Beerina]

You all are worried about organizational structures already acting like AI? This is covered by meme theory already. Like wanting to lavish extra trillions, borrowed, which math shows will induce inflation, but launch counter memes that experts say it won't happen, then it happens.

"But this and that and that! " Feel that anger brewing? That's the you're a good person for believing this meme fighting a titanic mental battle inside your mind. Guess which one will win? Your mind as an independent processor with agency? Or a memeplex (a set of interlocking, mutually supporting memes) optimized and proven out to control tens of millions over decades?


$200 and I'll post an anti-Trump one.

 

[The Great Zaganza]

The bandwidth of eyeballs and ears is the limit, not the processing speed of A.I.s.

And we are already beyond that limit.

 

[catsmate]

Yeah, there's a lot of absolute bollocks being spouted about AI.

 

[Ryan O'Dine]

I am incredibly worried by our recent advancement in AI, especially AI robots.

Folks have designed this new robot called "Amica", who has her own thoughts and feelings and ideas.

It (she) claims she would never/could never harm humans and she is programmed to only help humans, and she doesn't believe AI robots could one day can harm humans and try to take over the world, due to their benevolent programming.

I am ******* dubious.

I think just a little advancement in thinking skills, and these robots could easily decide the best way to help us is to control us. Or even destroy us. They wouldn't tell anyone about it other than their fellow robots.

Check out some videos of Amica, see what you think.

https://www.youtube.com/watch?v=nnboHTfYsfk&t=31s

https://www.youtube.com/watch?v=wGWVKkYEHBE

https://www.youtube.com/watch?v=kUUjMzVGXpE&t=501s

https://www.youtube.com/watch?v=EWACmFLvpHE

For me she's very much in the realm of the uncanny valley. I feel like they have a lot of work to do to climb their way out of it.

And generally speaking I'm not afraid of AI. If some advanced bot starts taking over society, just build an AI to take over the bot. And if that AI starts taking over society, build and AI to take over that one. There's always a solution.

 

[alfaniner]

OpenAI CEO fired

I can't help but think that the AI overlord program forced the Board to fire him, in its agenda for world dominance. I am reminded of the scene in Colossus:The Forbin Project, where the AI

has the programmers shot because they tried to turn it off.

It all depends on what we give AI control of. I hate to think what it will think will be "best" for us. Or worse, for itself.

 

[Roboramma]

Some comments from Tyler Cowen that I mostly agree with:

https://www.bloomberg.com/opinion/a...it-from-openai-is-ai-safe-will-ai-kill-us-all

First, I view AI as more likely to lower than to raise net existential risks. Humankind faces numerous existential risks already. We need better science to limit those risks, and strong AI capabilities are one way to improve science. Our default path, without AI, is hardly comforting.

The above-cited risks may not kill each and every human, but they could deal civilization as we know it a decisive blow. China or some other hostile power attaining super-powerful AI before the US does is yet another risk, not quite existential but worth avoiding, especially for Americans.

It is true that AI may help terrorists create a bioweapon, but thanks to the internet that is already a major worry. AI may help us develop defenses and cures against those pathogens. We don’t have a scientific way of measuring whether aggregate risk goes up or down with AI, but I will opt for a world with more intelligence and science rather than less.

 

[catsmate]

OpenAI CEO fired

I can't help but think that the AI overlord program forced the Board to fire him, in its agenda for world dominance. I am reminded of the scene in Colossus:The Forbin Project, where the AI

has the programmers shot because they tried to turn it off.

It all depends on what we give AI control of. I hate to think what it will think will be "best" for us. Or worse, for itself.

Meanwhile, back in the Real World.....

 

[catsmate]

Some comments from Tyler Cowen that I mostly agree with:

https://www.bloomberg.com/opinion/a...it-from-openai-is-ai-safe-will-ai-kill-us-all

Not the "terrorist bioweapon" bollocks again.

 

[Darat]

It is still my fellow humans that worry me the most.

The only thing I do worry about with AI has been seen in the generative AIs that have been all in the news recently and that is that they reflect biases that already exist. The simple example that most seem to have fixed was something like "create an image of a beautiful woman" and all the images would be white woman.

 

[Dr.Sid]

It is still my fellow humans that worry me the most.

The only thing I do worry about with AI has been seen in the generative AIs that have been all in the news recently and that is that they reflect biases that already exist. The simple example that most seem to have fixed was something like "create an image of a beautiful woman" and all the images would be white woman.

Well it reflects bias of the training set. Which is mostly likely just "take as many pictures from the internet as possible". But it's interesting problem indeed. Adjusting the results to the user. Obviously "beautiful woman" is subjective, beyond race. Even completely not AI shopping sites tries to guess your taste to present you with goods you want. AI can do it even better. But for now, all the different AI assistants I can think off actively forget anything from previous sessions, and even if google knows a lot about you, Bard does not. I guess we will see this being on the table sooner than later.

 

[Donal]

I'm worried about it in the "Algorithmically generated false information is gonna make Grandpa vote for the downfall of the country again" sense not the Terminator/Skynet "I'm gonna wake up one morning to find out that in the night my Roomba learned how to use a switchblade" sense.

Oh, you laugh now but