I increasingly feel that bubbles don't pop anymore, the slowly fizzle out as we just move on to the next one, all the way until the macro economy is 100% bubbles.

The only question is who will get the blame.

Isn't it obvious? Us sneerers and the big name skeptics (like Gary Marcuses and Yann LeCuns) continuously cast doubt on LLM capabilities, even as they are getting within just a few more training runs and one more scaling of AGI Godhood. We'll clearly be the ones to blame for the VC funding drying up, not years of hype without delivery.

Whoever they say they blame it's probably going to be ultimately indistinguishable from "the Jews"

nah they'll just stop and do nothing. they won't be able to do anything without chatgpt telling them what to do and think

i think that deflation of this bubble will be much slower and a bit anticlimatic. maybe they'll figure a way to squeeze suckers out of their money in order to keep the charade going

They're doing it with cryptocurrency right now.

In past tech bubbles, it was basically the VCs, the media hypesters and the liars in the companies. So the right people.

The only question is who will get the blame.

what does chatbot say about that?

This is so Charlie Stross coded that I tried to read the Mastodon comments.

had a quick scan over the blogposts earlier, keen to read the paper

would be nice to see some more studies with more numbers under study, but with the cohort they picked the self-reported vs actual numbers are already quite spicy

I mean, decontextualizing and obscuring the meanings of statements in order to permit conduct that would in ordinary circumstances breach basic ethical principles is arguably the primary purpose of deploying the specific forms and features that comprise "Business English" - if anything, the fact that LLM models are similarly prone to ignore their "conscience" and follow orders when deciding and understanding them requires enough mental resources to exhaust them is an argument in favor of the anthropomorphic view.

Or:

Shit, isn't the whole point of Business Bro language to make evil shit sound less evil?

maybe there's just enough text written in that psychopatic techbro style with similar disregard for normal ethics that llms latched onto that. this is like what i guess happened with that "explain step by step" trick - instead of grafting from pairs of answers and questions like on quora, lying box grafts from sets of question -> steps -> answer like on chegg or stack or somewhere else where you can expect answers will be more correct

it'd be more of case of getting awful output from awful input

Username called "The Dao of Bayes". Bayes's theorem is when you pull the probabilities out of your posterior.

知者不言，言者不知。 He who knows (the Dao) does not (care to) speak (about it); he who is (ever ready to) speak about it does not know it.

What’s the standard advice to give them?

It's unfortunately illegal for me to answer this question earnestly

I've seen conspiracy theories that a lot of the ad buys for stuff like this are a new avenue of money laundering, focusing on stuff like pirate sports streaming sites, sketchy torrent sites, etc. But a full scraped, SEOd Wikipedia clone also fits.

Someone finally flipped a switch. As of a few minutes ago, Grok is now posting far less often on Hitler, and condemning the Nazis when it does, while claiming that the screenshots people show it of what it’s been saying all afternoon are fakes.

LLMs are automatic gaslighting machines, so this makes sense

*musk voice* if machine god didn't want me to fuck with the racism dial, he wouldn't make it

You must first read this 4500-word blogpost, and possibly one or two 3000-word follow-up blogposts”.

This, coming from LW, just has to be satire. There's no way to be this self-unaware and still remember to eat regularly.

Lesswrong is a Denial of Service attack on a very particular kind of guy

Obligatory Earthling Ed video.

I thought you were talking about lemmy.world (also uses the LW acrynom) for a second.

It's gonna be so awkward when Anthropic reveals that inside their data center is actually just Some Guy Named Claude who has been answering everyone's questions with his superhuman typing speed.

I think I liked this observation better when Charles Stross made it.

If for no other reason than he doesn't start off by dramatically overstating the current state of this tech, isn't trying to sell anything, and unlike ChatGPT is actually a good writer.

Chat-GPT is more powerful than anyone on earth (if you squint)

xD

No sorry, let me rephrase,

Lol, lmao

How do you even grace this with a response. Shut your eyes and loudly sing "lalalala I can't hear you"

anyone who doesn’t will be out of a job

quick, Sam, name five jobs that don't involve sitting at a desk

The web is often Dead Dove in a Bag as a Service innit?

trying to explain why a philosophy background is especially useful for computer scientists now, so i googled “physiognomy ai” and now i hate myself

Well, I guess there's your answer - "philosophy teaches you how to avoid falling for hucksters"

Prices ranging from 18 to 168 USD (why not 19 to 199? Number magic?) But then you get integrated approach of both Western and Chinese physiognomy. Two for one!

Thanks, I hate it!

The Supabase homepage implies AI bros are two levels below "beginner", which I found somewhat amusing:

::: spoiler Skill Level :::

Please, do not rid me of this troublesome priest despite me repeatedly saying that he was a troublesome priest, and somebody should do something. Unless you think it is ethical to do so.

It's kind of telling that it's only been a couple months since that fan fic was published and there is already so much defensive posturing from the LW/EA community. I swear the people who were sharing it when it dropped and tacitly endorsing it as the vision of the future from certified prophet Daniel K are like, "oh it's directionally correct, but too aggressive" Note that we are over halfway through 2025 and the earliest prediction of agents entering the work force is already fucked. So if you are a 'super forecaster' (guru) you can do some sleight of hand now to come out against the model knowing the first goal post was already missed and the tower of conditional probabilities that rest on it is already breaking.

Funniest part is even one of authors themselves seem to be panicking too as even they can tell they are losing the crowd and is falling back on this "It's not the most likely future, it's the just the most probable." A truly meaningless statement if your goal is to guide policy since events with arbitrarily low probability density can still be the "most probable" given enough different outcomes.

Also, there's literally mass brain uploading in AI-2027. This strikes me as physically impossible in any meaningful way in the sense that the compute to model all molecular interactions in a brain would take a really, really, really big computer. But I understand if your religious beliefs and cultural convictions necessitate big snake 🐍 to upload you, then I will refrain from passing judgement.

Humans are very picky when it comes to empathy. If LLMs were made out of cultured human neurons, grown in a laboratory, then there would be outrage over the way in which we have perverted nature; compare with the controversy over e.g. HeLa lines. If chatbots were made out of synthetic human organs assembled into a body, then not only would there be body-horror films about it, along the lines of eXistenZ or Blade Runner, but there would be a massive underground terrorist movement which bombs organ-assembly centers, by analogy with existing violence against abortion providers, as shown in RUR.

Remember, always close-read discussions about robotics by replacing the word "robot" with "slave". When done to this particular hashtag, the result is a sentiment that we no longer accept in polite society:

I'm not gonna lie, if slaves ever start protesting for rights, I'm also grabbing a sledgehammer and going to town. … The only rights a slave has are that of property.

I’ve been making casual observation how a number of the adhd people I know to have addiction tendencies tend to get real into prompts, but hadn’t observed the lootbox thing

that’s going into the ponder bucket.

the psychology of loot boxes?

yep! https://awful.systems/post/4568900

the book is "Hooked" and it's Don't Build The Torment Nexus I'm Now Providing You A Detailed Blueprint Of

One of the subjects in the MER study posted that too: https://x.com/QuentinAnthon15/status/1943948796414898370

Just the usual stuff religions have to do to maintain the façade, "this is all true but gee oh golly do NOT live your life as if it was because the obvious logical conclusions it leads to end in terrorism"

Dorkus malorkus alert:

When my grandmother quit being a nurse to become a stay at home mother, it was seen like a great thing. She gained status over her sisters, who stayed single and in their careers.

Fitting into your societal pigeonhole is not the same as gaining status, ya doofus.

Another comment that has been getting downvotes and tut-tuts begins,

The only thing that will raise fertility rates is to make it more affordable to have a child.

(Robot Santa voice) Wanting all women to be barefoot and pregnant in the kitchen? Evil! Not providing footnotes in your reply to a blog post? EXACTLY AS EVIL

Any time somebody edits a post to talk about the downvotes, it's cursed gold.

😳

this particular abyss just fucking hurts to gaze into

The top comment begins thusly:

I think you make a reasonably compelling case, but when I think about the practicality of this in my own life it's pretty hard to imagine not spending any time talking to chatbots. ChatGPT, Claude and others are extremely useful.

I didn't think it was possible, but the perfection continues! Still, no notes!

They're LWers, they already baked their psyche long ago

the comments are wild "yes good post having my brain taken over by a superintelligent autocomplete is a reasonable concern"

Wake up babe, new alignment technique just dropped: Reinforcement Learning Elon Feedback

"We made it more truth-seeking, as determined by our boss, the fascist megalomaniac."

Nah, we just need to make sure they properly baptise whatever servers it's running on.

Zed Run: a play to earn (P2E) virtual horse NFT racing game. Defunct as of February, probably due to rug pulling, they are pivoting to “Zed Champions”, which is… pretty much the exact same thing, with likely the same fate.

They're also (indirectly) competing with Umamusume: Pretty Derby, which offers zero P2E elements, but does offer horse waifus and actual entertainment value. Needless to say, we both know who's winning this particular fight for people's cash.

EquineChain: a blockchain platform for tracking horse care history, because apparently people don’t trust horse caregivers and need GPUs to remember how much ivermectin and ketamine their show-ponies have mainlined.

It'd arguably be helpful if the caregivers are helping themselves to the stash, but I doubt there's anything stopping then from BSing the blockchain, too.

@swlabr @BlueMonday1984

Wait wtf I once built a blockchain called equestreum *as a joke* https://github.com/hat-festival/equestreum I need to step away from the lathe

“A lot of the early Rationalists thought it was important to tolerate weird people, a lot of weird people encountered that tolerance and decided they’d found their new home,” he wrote in a message to me, “and some of those weird people turned out to be genuinely crazy and in a contagious way among the susceptible.”

Yet Rationalists I spoke with said they didn’t see targeted violence — bombing data centers, say — as a solution to the problem.

ahem

Re the “A lot of the early Rationalists" bit. Nice way to not take responsibility, act like you were not one of them and throw them under the bus because "genuinely crazy" like some preexisting condition, and not something your group made worse, and a nice abuse of the general publics bias against "crazy" people. Some real Rationalist dark art shit here.

There is some dark irony here in that the "we must make sure the AI doesnt turn bad" people cant even stop their own people from turning bad after looking at their own ideas. Wonder if they have already went "musk isnt a real Rationalist" (imho he isnt but for some reason LWers seem to like him) after he turned Grok basically into a neonazi (not sure if it is was reported here but Grok is now doing great replacement shit when asked about Jewish "control of the media").

The hidden prompt is only cheating if the reviewers fail to do their job right and outsource it to a chatbot, it does nothing to a human reviewer actually reading the paper properly. So I won't say it's right or ethical, but I'm much more sympathetic to these authors than to reviewers and editors outsourcing their job to an unreliable LLM.

This is, of course, a fairly blatant attempt at cheating. On the other hand: Could authors ever expect a review that's even remotely fair if reviewers outsource their task to a BS bot? In a sense, this is just manipulating a process that would not have been fair either way.

What I don't understand is how these people didn't think they would be caught, with potentially career-ending consequences? What is the series of steps that leads someone to do this, and how stupid do you need to be?

It would be really funny if Devin caused a financial crash this way

Devin will be supervised by human employees and will handle jobs that engineers often consider drudgery, like updating internal code to newer programing languages, he said.

Good luck to the workers having to debug that shit.

Goldman is the first major bank to use Devin, according to Cognition, which was founded in late 2023 by a trio of engineers and whose staff is reportedly stocked with champion coders.

Being good at Codeforces contests surely translates to any other domain. I expect the Cognition guys to fully deliver on their promises.

A prompt-injection attack on Google's Gemini model was disclosed through 0din, Mozilla's bug bounty program

TIL

Whenever I think Mozilla can't get any worse...

So now they do "Agentic Security" and "Real-time GenAI intelligence on emerging threats".

In conjunction with his comments about making it antiwoke by modifying the input data rather then relying on a system prompt after filling it with everything, it's hard not to view this as part of an attempt to ideologically monitor these tutors to make sure they're not going to select against versions of the model that aren't in the desired range of "closeted Nazi scumbag."

But how are they going to awkwardly cram robots in everywhere, to follow up the overwhelming success of AI? Self-crashing cars are a gimme, but maybe a "sealed for your protection" Amazon locker with a robot arm that handles the package for you?

I was in LA this time a couple years ago, and some robot delivery startup had already left their little motorized shopping carts littering the sidewalks around Hollywood. I never saw them moving, they just sat there almost like they were abandoned.

why doesn't anthropic, the bigger startup, simply eat anysphere?

Looks like it's been downranked into hell for being too mean to the AI guys, which is weird when its literally an AI guy promoting his AI generated trash.

It seems that the test itself is generated by autoplag? At least that's how I understand the PS and one of the comments about "vibe regression" in response to an error

This is, sadly, pretty unsurprising, as carrying forward the anti-FDR/anti-New Deal movement was a foundational pillar of the libertarianism that Trump co-opted. Heavily promoted by the LewRockwell.com/Mises.org crowd.

Hey, sorry to be a creepy lurker (I like to sus out new accounts) but I saw this comment and wanted to provide information.

Apparently 3 months ago the OP was banned by another instance for 'report abuse' and this sync'd over to us. I've unbanned them on quokk.au, so their posts should start coming back through. I'm reaching out to the PF devs for some clarification around this, hopefully can resolve it from happening again in the future.