Your opinion is borne from lack of knowledge about the topic you’re trying to discuss, similar to an anti-vaxxer attempting to discuss the dangers of mRNA vaccines.

Fix your lack of knowledge or don’t; that’s your call, but if you care about discussing this topic somewhat intelligently, you should probably put in some effort to understand it.

I said several things so how about you actually address them, instead of resorting to an insult and vaguely telling me to “do my research”?

Do you deny this latest wave of companies doing the exact same routine is a PR stunt, or that companies badly want regulatory capture, or what?

I meant no insult, truly. You just don’t have the minimum required knowledge to discuss this intelligently. There’s no shame in that; no one can know everything.

I can’t know the motives of anyone, but I can say it’s at least plausible that the people working on the frontier models want an excuse to slow down that doesn’t give an advantage to other countries and companies, and also doesn’t put them in the crosshairs of investors and board members.

You apparently still internalize LLMs as fancy Markov chains, which tells me that you don’t have a working grasp to discuss the topic, even at the laymen level we’re at.

These hacking events may be “PR Stunts”, but the “stunt” part would be leaving a known vulnerability as a path to the wider Internet instead of making it physically impossible to break out, not the hacking of other websites. No one credible is disputing that these models are capable of hacking.

I meant no insult

Uh-huh. You compared me to an antivaxxer , implying that people who disagree with you don’t believe in science or technology - the same reason you felt the need to throw in the “inevitability” cliche. Which I do not buy, and somebody claiming to understand the technology should never repeat.

Later, you say anyone who disagrees with you just feels “desperate” and “threatened." You’re really pulling everything from the “I am rational and you are neurotic” playbook aren’t you.

I can’t know the motives of anyone, but I can say it’s at least plausible that the people working on the frontier models want an excuse to slow down that doesn’t give an advantage to other countries and companies, and also doesn’t put them in the crosshairs of investors and board members.

It’s all the same song and dance. Dario Amodei screams “hold me back!” because he loves pretending his product is Great and Powerful.

Why do these developers never talk about real harms like the environmental damage they cause? How about the psychosis they are liable to fall into?

It’s pretty obvious they’re just doing marketing, knowingly or not. And these are smart people who should know better.

Computer science is still science. You distrust the experts in the field based on what? Something you saw on social media? Certainly not your own expertise, that much is clear.

Don’t move the goalposts. No one is claiming that “AI” doesn’t have downsides. That’s different than your original claim that they didn’t do what they appear to do.

I don’t suppose you happen to be in a field threatened by AI? In my anecdotal experience, the people that desperately want to pretend AI is all a scam are actually worried about the opposite, and their job. (A real concern I’m in no way trying to minimize.)

I can’t tell if you’re reading what I’m writing. I mention CEOs twice and you assert I’m not listening to “The Experts”.

No one is claiming that “AI” doesn’t have downsides.

You mean climate damage and psychosis?

Zero of the “experts” in your strained theory about looking for excuses to “pause” AI ever talk about them. It’s almost like they don’t want to imply their product is unworthy of all the hype, doomerism and fearmongering.

I do not consider the CEOs as experts, no.

You’re not making coherent arguments; it really is like discussing vaccines with an antivaxxer. Maybe take a step back and explain to me what your stance actually is? Bonus points if you learn just a layman’s level understanding of the technology before you post your response.

Since your first post, I’ve been challenging you to justify your thought-terminating cliches, but instead all you do is string together insults like

it really is like discussing vaccines with an antivaxxer.

and slightly less overt ones. You should retract all the personal attacks or at least retract your insincere “apology.”

And for your vague sourceless claims about “experts” need to actually provide sources. Tell me the cult leader Eliezer Yudkowsky isn’t one of your favored “experts.”

If you don’t like that comparison, I suggest you take a hard look at how you’ve approached this topic.

I will happily continue this discussion once I (and presumably you) understand what you’re actually arguing.

I don’t like your insincerity with the fake apology, and I don’t like the insincerity of refusing to back up your claims.

We can start with the simplest question, which I already asked: Do you consider Eliezer Yudkowsky to be one of your “experts”?

I don’t understand the question. Yudkowsky’s book is on my reading list, but judging by the title, I assumed it to be more on the side of restricting AI research, which best aligns with what I think (but suspiciously you refuse to clarify) your stance is.

When I said “the experts” I meant all the people working on AI at these frontier AI companies. You know, the experts?

I disagree with Yudkowsky in pretty much every way possible, and don’t know how you could conclude I agree with him if you saw anything else I posted in this thread. He’s like you, he believes AI is inevitable and wants it to be good too.

When I said “the experts” I meant all the people working on AI at these frontier AI companies

Interesting how you conflate “expert” with “employee” when determining who you trust. Do you see any conflict of interest here?

And I asked you earlier about why these “experts” never sound the alarm about the environmental or mental harms of AI. I wonder why that is… Do you think it hurts their bottom line?

I answered your non sequitur as a show of good faith. Your turn: what precisely is the stance you’re arguing here?

That AI doomerism is AI marketing.

Now I’m interested in how you conflate “expert” with “employee” when determining who you trust. Do you see any conflict of interest here?

And I asked you earlier about why these “experts” never sound the alarm about the environmental or mental harms of AI. I wonder why that is… Do you think it hurts their bottom line?

Slow down, sir or ma’am. We’ll get to your questions, I promise.

What does “AI Doomerism” mean? I am not familiar with the term.

https://ondiscourse.com/ai-doomerism-is-a-business-tactic/

Now I’m interested in how you conflate “expert” with “employee” when determining who you trust. Do you see any conflict of interest here?

And I asked you earlier about why these “experts” never sound the alarm about the environmental or mental harms of AI. I wonder why that is… Do you think it hurts their bottom line?

I know it has already been said, but even your tactics resemble anti-intellectualism. You don’t need to keep hammering the same questions; I will get to them. Calm down, there isn’t a prize at the end; there is no winner or loser. We’re just having a discussion. Breathe.

That link didn’t answer my question, so I have to ask more questions. Are you saying that AI will not (can not?) be a threat to humanity, ever? And is the rationale for that conclusion because you don’t think AI actually works?

Do you really think that your insults are less insulting if you follow them up with condescension? If you refuse to answer my questions, just say so, instead of being a douchebag.

I answered you twice, “in good faith” as you like to say, and it’s clear that’s something you’re not here in.

You haven’t answered my questions, hence why I still have questions. I promise you that, even if it’s the last comment I make to you, I will answer your questions. They’re not the gotcha you apparently believe them to be and are essentially a waste of time to answer, but they appear super important so I will get to them.

First I want to understand your stance. All I’ve got now is that a blog said AI isn’t actually dangerous, and you concur. Why do you think this?

Calm down, there isn’t a prize at the end; there is no winner or loser. We’re just having a discussion. Breathe.

You started this off claiming that this person is equivalent to an antivaxxer. You’ve repeatedly dismissed their claims/thoughts based on credentialism without demonstrating why. Now you’re telling them to calm down. I’m not even the person you’re arguing with here and I’m getting mad lol.

They are using the same playbook as antivaxxers. There’s no way around the comparison. It wasn’t meant as an insult, but as an observation. Notice they didn’t correct me, but instead indicated that it hurt their feelings.

I’m not arguing with anyone; they have not yet made a point that required more than a factual correction, or a point that was just a conspiracy theory.

You definitely shouldn’t get mad about it though, haha.

@joe @XLE "I don’t suppose you happen to be in a field threatened by AI?"

Living on Planet Earth is a good start, you arrogant little fart. Your carefully assumed air of superiority is the same as that of all the other scammers trying to sell snake oil that I've encountered over more years than you've been active, #ZeroFollowersFakeAccount

Ignoring the random personal attacks, I’m not on Mastodon like you are, I’m on Lemmy. We don’t have followers, that I am aware of.

Wow. What an awful person.

That cliche is the worst. If applied to anything else you can see how absurd it is. Like NFTs aren’t going anywhere, crime isn’t going anywhere, etc.

Except NFTs were never used to nearly even close to the same amount as AI, it had much less usecases and the one that did “catch on” was idiotic.

Crime is a symptom of a problem, not a technological advancement, apples and oranges.

AI the tool is never going away in the same way the internet is never going away. (Barring our entire civilization collapses and we lose knowledge/connectivity/life lmao)

“AI is never going away” neither did NFTs nor crime, nor anything else really. We should try to stop these awful companies from doing awful things.

I agree?

Then what’s the point of saying that line? It’s a pointless line that means to put down any solution to the problem saying “well, actually, it’ll never go away!” instead of thinking about how one could drastically reduce the harm by trying to make it go away anyways

Because the underlying problem isn’t the tool.

What is “the tool”, GenAI or ML. If you mean GenAI, then, no, GenAI is the problem

The point of the “its not going away” comment is because a shocking number of people think the solution to AI is to, I dunno, delete it.

As a technology, it is out in the world. It isn’t going away any more than electricity or the internet. So any solutions to the real problems with AI that are essentially “make it go away” are essentially nonsense. Might as well propose wishing it away with a genie.

“It’s not going away” is not equivalent to “there’s nothing we can do to solve the problems”; it just means that the solution has to be in a framework where it is implemented alongside AI.

That is a solution, make it very unappealing to use it. If it’s clearly a crime, most people don’t have the hardware to do it, and others don’t have the motivation, it’d help a lot. And again, NFTs

Most people do have the hardware to run them. Maybe not super high end ones but middle of the road ones.

Again, the solution won’t ever be “no more AI”. The solution will be regulation, social safety nets, taxation, etc.

They are crimes, they stole their data, and the ones people can run locally still are a part of those crime rings. Plus the ones that can run locally aren’t very good, not that the larger ones are too great either

What court case are you referring to?

What court case did I bring up?

How do you determine that a crime has been committed without a court case?

Same way we always do? look at what happened, and see that a crime happened

That breathtakingly wrong. Yikes.

Do we need to have a court case to say that all of the criminals in the files are criminals? of course not. We can see their actions, and the laws and say that they committed many many crimes.

Yes, you need a verdict from a court to know whether someone committed a crime. Is this not taught in school anymore??

You can’t just accuse someone of breaking the law and end it there. Holy hell.

you need schooling, this is like people who shout that others are breaking the first amendment not thinking that it only applies to the government. Courts determine if it’s a crime to a certain degree of certainty for the government, we are not the government. We are people who can use our eyes to see that people break the laws and have not been sued yet. it’s not magically not a crime if they aren’t sued over it.

So if someone on the internet calls you a thief and I see it, you’re a thief?

Think this through.

So in order to counter my point, you make a false equivalence. Cool. Try again.

Says the person who set up a strawman about what you imagine might say about the first amendment vs freedom of speech. lol

I’ll make this easy. An accusation is not evidence, let alone proof. That is not up for debate; it’s a fact. You can be accused of literally anything.

I tried to hold your hand and walk you there, but that clearly wasn’t working.

Would you even want it to work the way you did falsely believe it works? Do you want any allegation to be enough to label someone a criminal? Have you spent even 30 seconds considering the implications of this? Seems not.

We are not a court of law. We can see crimes happening and call them crimes. If you think that ignoring evidence is fine, that’s up to you. I personally like to look at the evidence and make conclusions, like these crimes AI companies are doing, are crimes. Did the AI companies think about for 30 seconds about how doing crimes looks like doing crimes?

If you’re not a lawyer I doubt you even know enough about the law to determine what counts as a crime, and you certainly don’t have all the evidence.

Here, I’ll toss you what should be an easy one: what crime?

Violation of copyright/piracy :P Among many many many other crimes. I’ll take the word of the US copyright office who says that it is not fair use over your word.

Make your case.

The US copyright office wrote up a pretty large paper pointing out it’s piracy.

Source?

Alignment is a real thing, though it can be a less than apt description. Mostly the goals you write out and the goals you want to happen not being the same thing. You can also have alignment issues where the dataset doesn’t match the real data in some way and the classifications break down. (I am not an AI defender I am very anti-AI)

I thought alignment was just preventing the llm from making porn or bomb recipes to avoid legal trouble

No, that would be bad, but not misaligned necessarily. That’s a different issue really. Like lets say you use ML to make an AI to play a video game and the goal is always on the right side, but after you’re done training, it’s on the left now, the AI might have not learned go the goal, but go to the right.