Adam Conover: AI Hoax Distroying America with Ed Zitron

[Christina]

Ok, wasn't sure where to post this. Move it if someplace else would be better. I care not.

This interview sums up my feels on the current state of AI, the economy, the rich tech dumbasses. What I think is the bubble that is about to burst and send us all into another even greater depression.

So I needed to share this with others. Feel free to use this thread as a place to vent out your frustrations on how in the world did we end up here. Like watching a slow moving train. This guy has made me feel not crazy for thinking all the things he is saying here.

 

[Kat]

AI as a technology is fine, but it's marketed and used in inappropriate ways. You cannot use AI to decide which people to fire, or to run a suicide helpline, or all the other dumb shit execs use it for. It's a tool, like anything. It doesn't help that they fed it Reddit, of course... It's unfortunate, really, because it's such a cool advancement from a scientific standpoint.

I really hope Elon quickly develops his rocket to Mars and takes all the other tech bros with him. Maybe they should bring some other execs with them too. Our world would be so much better without them.

 

[Christina]

AI as a technology is fine, but it's marketed and used in inappropriate ways. You cannot use AI to decide which people to fire, or to run a suicide helpline, or all the other dumb shit execs use it for. It's a tool, like anything. It doesn't help that they fed it Reddit, of course... It's unfortunate, really, because it's such a cool advancement from a scientific standpoint.

I really hope Elon quickly develops his rocket to Mars and takes all the other tech bros with him. Maybe they should bring some other execs with them too. Our world would be so much better without them.

When I heard Neil DeGrasse say, "if we could terraform Mars, we wouldn't need to because that would mean we could terraform Earth and fix Earth"

Musk is the idiot that tosses away billions of dollars trying to do the thing he saw in a video game once.

 

[Kat]

When I heard Neil DeGrasse say, "if we could terraform Mars, we wouldn't need to because that would mean we could terraform Earth and fix Earth"

Musk is the idiot that tosses away billions of dollars trying to do the thing he saw in a video game once.

Wait, Musk expects to terraform Mars?? Lolol god that guy never ceases to amaze in the worst ways.

 

[Christina]

Wait, Musk expects to terraform Mars?? Lolol god that guy never ceases to amaze in the worst ways.

I honestly think he gave up on the Mars thing when he realized how hard it would be. He had to stop and bail himself out of all the investigations the government was starting to do. He also needed more money from the government because he ran his mouth about the person giving Space X most of it's money.

Here's a 10 minute broad overview of it. Elon's companies have always been overvalued and they have not been doing as well as their PR campaign is always trying to make it out.



This isn't just an Elon thing. The whole tech sector is way overvalued and is always losing money for years now. Especially with all the focus on AI which is a horrible money sink.

 

[Mark]

I don’t have anything insightful to contribute yet, but, remember SmarterChild? That’s when I knew that we as a species weren’t responsible enough for any form of artificial intelligence.

 

[Christina]

I don’t have anything insightful to contribute yet, but, remember SmarterChild? That’s when I knew that we as a species weren’t responsible enough for any form of artificial intelligence.

Omg even then they couldn't find a way to make it a profitable thing that anybody wanted or needed. You would think they would have learned their lesson, but nope.

 

[Mark]

Omg even then they couldn't find a way to make it a profitable thing that anybody wanted or needed. You would think they would have learned their lesson, but nope.

I’m convinced that SmarterChild and other similar bots of that generation were used to compile hard data on the human input it received. They weren’t even remotely profitable, but, they collected plenty of information on the kind of absurd questions people would ask it, which would prove to be extremely useful in later iterations.

 

[Nae'blis]

Ever since Musk started selling the flamethrowers, I've been getting mad Hank Scorpio vibes from him.

 

[Crazy Jamie]

This thread very much ties into my thread about the Enshitiffication of the Internet. Ed Zitron has a podcast called Better Offline, which deals with this and other related issues. I stopped listening because it was very repetitive and it really is just him ranting, but the points he made were clear and persuasive. AI is a favourite topic for him, and when I actually read into it he seems to be to be absolutely right. AI can do a great many things very well. But as Kat says, the problem is that it's being marketed as a revolution for everything. It isn't. There are also very many things it cannot do. At all. It's been naturally incorporated into several programmes that I use, such as Outlook and Mail, and it is absolutely worthless. At least for me. But its capabilities are obviously being massively inflated because, as is said at the start of this podcast, it's ultimately about money, and the people saying it can do everything are the ones becoming obscenely rich through everyone believing it can do everything.

Unfortunately it's part of an ongoing trend, which includes the likes of BP and Shell confirming that they're going to cut back on renewable ventures because they're not making as much money. That just encapsulates all of this perfectly. The money makers have massive influence over government and policy, and even outside of that. And it's all about money. Any of them suggesting that they're working for any sort of greater good is entirely disingenuous. And when these things burst, because they do, it's normal people that suffer. It's a cycle that doesn't end.

 

[Kat]

It's been naturally incorporated into several programmes that I use, such as Outlook and Mail, and it is absolutely worthless.

It's been added to Visual Studio too and the code suggestions it generates are garbage. It suggests the wrong variable type even when the right one is obvious (as in the compiler can figure it out even if you don't specify), suggests class members that don't exist, etc. If you're making the exact same type of change a bunch of times in a row, occasionally it'll figure it out, but after three or four of them it suggests something random again. It's easier just to type the right code from scratch IMO.

IMO that's annoying but less of an existential problem than putting AI nonsense at the top of every Google search or using it to decide who to lay off. If AI suggests nonsense for an email or the code I'm writing, it's obvious it's nonsense. It's wasting some clock cycles on my computer and a bit of brainpower to read it, but that's it. It's when people use it to answer a question and they cannot easily check its correctness (or might not realize they need to) that it's a real issue.

If it ever becomes actually effective, we're going to have some serious problems with the job market, but I'm not expecting that'll happen any time soon. They've run out of data to feed it.

 

[Christina]

If you pretend for a moment that they did make artificial human level of intelligence (which they did not and we don't even know how to define that anyways). Humans lie. We lie and for all kinds of dumb reasons. We lie because we just want to sound smart. We lie because we thought we knew the right answer. We lie because it's funny. We lie just because sometimes and we don't even know why.

I don't want a human search engine. I want a dumb machine search engine that just knows how to spit out whatever facts it's been taught.

 

[Kat]

Yeah, I agree with that, Christina. The idea of AI in science fiction is that it's always accurate, doesn't lie, is reliable and consistent and predictable, etc. What a shock that when we try to replicate our intelligence, it ends up acting like us instead.

 

[Christina]

Yeah, I agree with that, Christina. The idea of AI in science fiction is that it's always accurate, doesn't lie, is reliable and consistent and predictable, etc. What a shock that when we try to replicate our intelligence, it ends up acting like us instead.

I thought the matrix was the only sci-fi that did really well with this. Go figure that trans people have that worked out better.