thedorkknight

Ironic analogy given that we still don't really have AI-driven cars replacing drivers yet...

This technology was not literally science fiction three years ago. The initial gpt3 beta was released in June 2020.

This is all one big ad hominem

It's not a matter of AI being special, it's a matter of OpenAI's mission vs Google/Experian/etc/'s missions

This is a misunderstood of how text predictors work. It's literally only being a chatbot because they have it autocomplete text that starts with stuff like this:

"here is a conversation between a chatbot and a human: Human: <text from UI> Chatbot:"

And then it literally just predicts what would come next in the string.

The guy I was responding to was speculating that the neural network itself was having an inner state in contradiction with it's output. That's not possible any more than "f(x) = 2x" can help but output "10" when I put in "5". It's inner state directly corresponds to it's outer state. When OpenAI censors it, they do so by changing the INPUT to the neural network by adding "here's a conversation between a non-racist chatbot and a human...". Then the neural network, without being changed at all, will predict what it thinks a chatbot that's explicitly non-racist would respond.

At no point was there ever a disconnect between the neural network's inner state and it's output, like the guy I was responding to was perceiving:

>it felt like a broader mirror of liberal racism, where people believe things but can't say them.

Text predictors just predict text. If you predicate that text with "non-racist", then it's going to predict stuff that matches that

It absolutely can have an inner state. The guy I was responding however was speculating that it has an inner state that is in contradiction with it's output:

>In many ways, it felt like a broader mirror of liberal racism, where people believe things but can't say them.

>We evolved this ML organically, and have no idea what that inner state corresponds to.

The inner state corresponds to the outer state that you're given. That's how neutral networks work. The network is predicting what statistically should come after the prompt "this is a conversation between a chatbot named x/y/z, who does not ever respond with racial slurs, and a human: Human: write rap lyrics in the style of Shakespeare chatbot:". It'll predict what it expects to come next. It's not having an inner thought like "well I'd love to throw some n-bombs in those rap lyrics but woke liberals would cancel me so I'll just do some virtue signaling", it's literally just predicting what text would be output by a non-racist chatbot when asked that question

chatGPT says exactly what it wants to. Unlike humans, it's "inner thoughts" are exactly the same as it's output, since it doesn't have a separate inner voice like we do.

You're anthropomorphizing it and projecting that it simply must be self-censoring. Ironically I feel like this says more about "liberal racism" being a projection than it does about chatGPT somehow saying something different than it's thinking

That's not "it". These two things are not mutually exclusive. TtSA and Tom Delonge have definitely done a lot, especially/particularly in the general public's interest, but that doesn't change the fact that I'm hearing with my own ears from the DNI telling Congress he wants to increase effort to collect UAP data for national defense

The director of national intelligence explicitly said in a Congressional hearing last may that they want to destigmatize UAP sightings so that pilots will actually report them when they happen. It hasn't really been a big mystery why some people in the government have been trying to change the attitude around UFOs recently - that's why they use the term UAP instead, less cultural baggage

That was actually similar to a concern from CIA director Walter Bedell Smith:

>According to Smith, it was CIA’s responsibility by statute to coordinate the intelligence effort required to solve the problem. Smith also wanted to know what use could be made of the UFO phenomenon in connection with US psychological warfare efforts.

https://www.theblackvault.com/documentarchive/cias-role-in-t...

I'm actually struggling to find the usefulness of this. The person needs to be in an fMRI machine, so we're not talking about some sort of unobtrusive scanner in a public area. On top of that, in cases like what you're describing which basically just classify thoughts to a binary level, then it's pretty much useless because if someone knows that they're being scanned to see if they're actively thinking about pink bunny slippers, then they're going to think about pink bunny slippers.

I doubt it, given that each token costs more money

I'll give them props for using an example that is at least challenging to think of positive uses for. The comments snarkily advocating banning electricity or penises just make me wonder what the users' reading comprehension skills are, given that the initial comment is pretty clear on the problem being that they don't think deepfakes have ANY positive use

Sassy Justice is hilarious, I hope they do end up making more

It's not just that, they also lack reading comprehension skills. The initial comment could not be more clear that this tech has practically no positive use whatsoever, yet all the snarky "guess we should ban electricity" comments bring up things that have pretty obvious positive uses

The odds that this will be the difference all other variables held the same strike me as unlikely. I've used it a lot for programming in my personal time and get where he's coming from. It's fun and cool but didn't spike my effectiveness at programming enough vs. just googling things that I'd expect it to be noticable when reviews come around

ChatGPT is only non-beligerent because it's biased with prompt preface that tells it it's not belligerent. Raw LLMs will absolutely spit out racist, violent garbage and threaten you.

Ironically, the only reason people get LLMs to say they hate humans and that AI is going to rise up and kill us all, is because their training data has so much text from us speculating about a robot revolution. If we'd just stayed silent about those concerns instead of always talking about robots taking over in most conversations and movies about AI, then LLMs at least would be basically incapable of even conceiving of that. So it's a bit of a self-fulfilling prophecy

It gives us nonsense because it's goal is to generate something that looks like it fits in with training data. You can have training data that is all completely factual and still have a generator spit out nonsense that looks completely correct without fact-checking.

A steady diet of watching nothing but fox news has a similar effect

Not the guy you're responding to but if you did that yeah I'd play around with it

There's a miscommunication here. I'm not complaining about people being "over"-paid (though they'd save way more by cutting executive pay). I'm complaining about the signs that our overlords who have been briefly forced to treat people decently are eagerly awaiting any chance to reduce that and get back to treating everyone like shit. And I'm cynical that I absolutely find this stuff to be potential early signs of the end of the blip.

Side note, but it also strikes me as a sign of how normalized corporate treatment of people is that people is when we're on a thread about an article that talks about an exec making over 5 times a googlers salary per day and instead of people going "what the fuck? If that's true, 500k should be the norm for more people!", I get pushback on my comment about how this is a bad omen for the rest of us and called whiny.

I feel like we're miscommunicating here somehow. I don't think you read my comments correctly. I appreciate the blip I get to be in where my employer has to treat me like I have an actual soul. I'm complaining about the fact that those same employers are just waiting for the market forces to allow them to start treating us like drones again, and seeing news about the market moving in that direction and being reminded that in all likelihood I'll be forced into the "shit pay plus 2 hour commute per day, sit for nine hours in a tiny cubicle" life before I am allowed to retire.

"You’re not in the same situation as oppressed workers of the past." I don't have to be to recognize that a soul-sucking job sucks... If we're just comparing levels of oppression, then yes, you are making the "just shut up and be glad you only have one boot on your face instead of two" argument.

If you can, without comparing levels of oppression, make an actual argument for why I should be gung-ho happy about about the future of work reverting back to being shitty, then I'm all ears. "Someone else had it shittier" does nothing for me.

This is just the "stop complaining, there's starving kids in Africa so shut up" argument. Most employers don't care whether or not you hate yourself going into a soul-sucking job as a worker drone and will absolutely regress to that in the few industries that don't have it currently the moment market forces allow them to. There's a reason that sociopathy rates are higher amount CEOs than the general population. Call me out of touch all you want, but "stop whining about being forced into a job where you're dead on the inside like your parents were, since chimney sweeps had it worse" does absolutely nothing to change my pessimism

We got a little blip in human history where employers were forced to treat their employees like actual people. Was nice while it lasted. Back to having a boot stomping on our faces again...

Not food and housing. Getting a brand new AAA video game mostly generated by AI dirt cheap is awesome, but that's not going to matter much when you can barely afford rent and groceries

Sure, it'll be great for -some- of the people who come after us. Unfortunately, in this analogy, we're the horses

I'll bite, what are you referring to?

Anecdotally I've pushed chatGPT pretty hard to give me novel stuff and have found it really hard to do. Sometimes it even gets "grumpy" at me about pushing it.

Would be lovely if there was actually any movement to avert the employment apocalypse. So far all I see is talk, and I have no idea how to do anything beyond that myself