(I get the feeling we’re close to this bubble’s end.)
Anyways, welcome to the Stubsack, your first port of call for learning fresh Awful you’ll near-instantly regret.
Any awful.systems sub may be subsneered in this subthread, techtakes or no.
If your sneer seems higher quality than you thought, feel free to cut’n’paste it into its own post — there’s no quota for posting and the bar really isn’t that high.
The post Xitter web has spawned soo many “esoteric” right wing freaks, but there’s no appropriate sneer-space for them. I’m talking redscare-ish, reality challenged “culture critics” who write about everything but understand nothing. I’m talking about reply-guys who make the same 6 tweets about the same 3 subjects. They’re inescapable at this point, yet I don’t see them mocked (as much as they should be)
Like, there was one dude a while back who insisted that women couldn’t be surgeons because they didn’t believe in the moon or in stars? I think each and every one of these guys is uniquely fucked up and if I can’t escape them, I would love to sneer at them.
(Credit and/or blame to David Gerard for starting this.)
The second affair has hit Nvidia stock
Sorry I have to vent, but looks like the humblebundle site did an AI powered redesign as the site (while looking like it works) is now a total mess. I’m a bit in slow motion trainwreck watching mode. Sucks I do have a list of unclaimed keys still at the site which I paid for (some of which are now gone, as in I know I still had a key as I kept a list of some of the keys in a file (I was worried something like this happening for a while) but the keys are no longer in the list of games I got, this is just one of the many many bugs or just plain weird things I noticed (The reddit is filled with people having a lot more complaints, somebody bought a ‘4 for 1, play with your friends’ pack, and only got one key, if you made a ‘gift a key link’ yday the link now is dead, but the game counts as claimed, etc etc).
The UI does have nice rounded corners now however.
E: additional funny moment, so I know people who have subscribed to humble choice (a monthly thing) an basically have forgotten about it, but during the creation of the new site they messed something up and they no longer have the information on file to complete the transaction for that, so a lot of subscriptions will likely auto cancel, what a foot shot).
Ah, the dev on main and push on Friday school of development.
Complaint driven development if you will.
God bless linkedin lunatics
Transcription of the image below
People often forget that Jensen is the hero who brought the light of modern AI to every household. He is also one of the most technical and charismatic CEOs in tech. He single handedly contributed more to the US and global economy than 80% of all other tech execs combined. Jensen is the heart and mind of Nvidia and one small mistake on the personal level only shows that he is human. Do not judge. ♥️ Jensen ♥️
Yes just a small personal mistake like blowing up your marriage and committing treason 😂😂😂
Treason is cool and good. Cheaters can walk into a fire.
calling what jensen did ‘treason’ accepts far too much of both the AI boosters’ and the corrupt imperialist US government’s worldview, in my opinion
yeah 100% but like he could have rounded up all us politicians into one room and sold the GPS coordinates to China who then bombed it and it’d still be awesome.
You can’t betray people that impose themselves on. Treason is fake bullshit
The Git project has put its current LLM policy (which currently blocks some AI use) up to a vote. An easily manipulated vote, like the kind Debian did. For the version control software used by the vast majority of software projects across the world.
Why don’t they just vibe code their own implementation of anthropic Scion and leave git to us “luddites” - especially since git isn’t fit for purpose any more when you’re slopping out 30k changes a day
Cathartic venting, tangential to the ordure du jour: in high school everyone I had a crush on had a crush on Joseph Gordon Levitt. It brings me great joy to know that he is now alignment brainrotted, and I am not.
The AI 2027 authors are back: Q3.5 Timelines Update: Looks Like the Intelligence Explosion Will Happen in 2027 After All
It’s all about that new METR data. Took a quick scroll and the whole thing is about forecasting and modelling. Its pretty lengthy so if anyones willing to take the plunge and read it all youll have my thanks
METR recently estimated that Anthropic is now making AI progress about twice as fast as its historical rate (they would not be surprised by anything from 1.3x to 3x), and guessed that “perhaps ≈80%” of the current acceleration is due to AI. If they’re right, that implies an AI-driven acceleration that is 5 months ahead of the AI 2027 scenario.
Going down a little bit:
A 1.7x pace of progress is 5 months ahead of the AI 2027 scenario. AI 2027 predicted 2x AI R&D uplift in Dec 2026; as argued above, this would imply a substantially lower pace of progress than the 1.7x AI-attributable pace of progress that METR’s estimates imply for Nov. So when exactly did AI 2027 predict a 1.7x pace of progress over baseline? We created this trajectory by modifying a few key parameters until it roughly matched AI 2027. In that trajectory, 1.7x acceleration over the 2024.0–2025.5 rate of progress happens in roughly April 2027. So by this metric, we are ~5 months ahead of schedule.
And a chart of their predictions to end it all off:
(Only one person above 50%?)
Mark your calendars everybody!
yeah but wen fuckable digital jesus.
I have had to explain to my boss that METR are cultists
I vaguely remember it being mentioned that one of the METR staff was married to a rationalist (it was either an Anthropic employee or Paul Christiano, I forgot who)
It’s all about that new METR data.
Yeah, they seem to have moved on from multiplying sufficiently large matrixes is magic to thinking it’s actually the code writing that’ll get mystical any time now.
One more if statement bro
TED-AI as in the meaninglessly distilled and contrived ted-talk idea of AI
Broke: “it is AGI when it makes A BILLION DOLLARS”
Woke: “it is AGI when it does a ted talk”
by that definition we already have AGI. Not because it’s that good but rather because TED was already brainrot slop well before LLMs.
If AI will not come closer to AGI, the definition of AGI will come closer to AI.
Pretty sure its just a fancier term for AGI (Top Expert Dominating AI)
“Top Expert Dominating” is when someone with a lot of experience makes someone else discover that they’re a switch after all.
I intended it as a sneer lol. Aka I don’t see a difference in meanings here
ohhhh
So I went all the way down to the appendix, and found a gem:
Appendix: Ambiguity in “80% of current acceleration is due to AI”. Claude Opus 5.5 gives the following possible interpretations of METR’s claim that “we guess that a majority (perhaps ≈80%) of the current (October 2026) acceleration is due to AI.”
Claude Opus 5.5 gives the follow possible interpretations
Did…did they just say they used Claude for this??
For years the LW types have just been saying ‘i asked claude’ about EVERYTHING
Wholesome indie Darling about human connection actually slopgobbled https://aftermath.site/big-walk-ai-igf-jurors-controversy/
discloses now rather than at launch because they’re just so wholesome and committed to ethics.
Arguing over homeopathic doses of AI misses the point like with actual homeopathy. They used it.
Developers House House told Aftermath via email that “LLMs were used sparingly by House House to help work through difficult technical challenges and bugs”, and that they “estimate that well under 0.1% of Big Walk’s code had anything to do with this usage”. They also stress that “generative AI was not used in any other context when making the game”.
Not to get all doctorow on this, but I generally don’t think we should concede this approach to using LLMs to the yooooooo I just vibecoded an FIFA 2027 cloen for android watch WITH MY DICK!!!11!! people.
If/when a time comes when genAI isn’t actively ruining everything and doesn’t need all the RAM and hard drive space in the world to be useful, using it like that seems a far cry better than even what torvalds is currently doing.
Build a machine that isn’t made of trust violations, controiled by the worst people in the world, and actively making climate change worse potentially killing millions first slopper.
Sure seems that AI usage more and more has the stink of ‘you can trust people who use this at all’.
Unrelated to that this is also very odd:
Then, we were able to speak directly to an engineer at Unity, who asked around and ultimately concluded that what we wanted wasn’t possible.
Unity jumped on the slop train some time ago, so I’m chalking that up to the codebase being flooded with slop.
What advantage does unity have over godot these days anyway?
I feel like I’ve seen more than one person around the Interwebs make the joke, “It’s a brutal betrayal to discover that Big Walk uses Unity.”
Nevermind my suggestion, of course they also use AI shit now.
I’ve been waiting for this
top quality sneer from “@1techuptopiapls” on X the fascism app:
”man proves AI improvements are not economically viable as new Anthropic model successfully cures cancer"
is the cure to cancer in the room with us right now
A number of responses (like, 2 out of the 3 that it shows) refer to ‘the scaling laws’ without elaboration. Is this some jargon special to AI bros?
yeah essentially
the scaling laws are basically that AGI/ASI can be created by just scaling up the current LLMs
But they really aren’t! The actually interesting scaling laws from about 5-7 years ago that opened up machine learning to more applications all show that everything about machine learning is power-law diminishing returns with linear improvements in any metric requiring exponential increases in numbers of parameters / amounts of data / number of backprop cycles applied.
uh but have you considered unbounded recursive self-improvement? it’s how it works in books and in my imagination
The scaling laws maxed out around mid 2023 with the release of GPT-4 - that was around the time that everyone got excited about “test time inference” which is where LLMs spit out reasoning streams to convince you they are “thinking”. If you look at where LLMs have improved since then it’s mainly in tasks where they can be strapped into a harness and given concrete feedback like “your program didn’t compile” which are also coincidentally use cases where data from gpt and Claude users can easily be harvested and new training data can be squeezed out of underpaid third world mturkers.
In other words they’ve improved their training data in the specific domains they are operating in like we’ve been doing in machine learning since the mid 90s…
The big labs have just been doing an incredible job of marketing (and particularly in Anthropic’s case, brain damaging themselves into believing their own marketing)
These scaling laws are even funnier than that, as are more recent ones.
I continue to be extremely amused by the fact that you can take a LLM ‘reasoning’ trace, replace all the ‘reasoning’ with dashes or spaces, re-run the attention mechanism, and have what comes out the other end improve by a significant fraction as much as if you let it exude text that has the appearance of having to do with the topic at hand. It’s just more opportunity for the embedding vectors to dance against each other before you translate the abstract vectors back into natural language. Hence recent frontier models doing a thing where instead of doing a long reasoning trace, they build the attention matrix, allow the vectors to push each other around… then let those new vectors push each other around again, and again, and again, multiple rounds, before you start decoding.
More recently, I have seen some vague math into the scaling of the speed with which running many parallel systems with access to a common pool of text produce a legible result. If I recall, it comes to roughly:
N parallel systems
Finish in ~1/N^0.5 as much time
At ~N^0.5x the cost
So 16 parallel systems crosstalking finish complex tasks in 1/4 the time at 4x the cost (and 16x the instantaneous rate of operation per second)
This guy probably thinks that outputting sloppily written proofs to math problems is the same thing as curing cancer because both of them are done by super smart scientific brain geniuses and the AI is an even bigger brain genius.
most of the other replies are all ‘well yeah we knew sustaining AI progress would need exponentially more resources but its like definitely worth it’
When children go to sleep, the Sandman shows them dreams. When AI boosters go to sleep, the Smiling Man shows them nightmares.
And that gets us back to istewart and Dave Karpf’s point that these companies sell a narrative about what the future will be and who controls it, because they can not build that future.
It’s just ‘well the financials don’t make any sense and these companies are not looking profitable anytime soon but like AI models are super duper good and AGI is right around the corner so you need to invest’
new odium symposium episode. we talk about Lawrence Mead, the NYU professor who advanced much of the theory behind the US welfare reform of the 90s, and his more recent project of inventing a newer, better racism. https://www.patreon.com/OdiumSymposium/posts/28-racism-2-174814488 or any podcast app
very gendered and heteronormative (and it admits it), but might be interesting?
Total System Failure: U.S. Government Networks Will Need To Be Torched And Rebuilt
Interesting if you want a rundown of US gov IT security fuckups, so far, in light of nvidia’s Jansen being recently revealed to have been compromised by Chinese intelligence.
I’ve noted repeatedly in my writing that the Trump administration’s cybersecurity policy has been indistinguishable from a foreign attack.
I have wondered for a long time what sorts of clandestine messages are going over the newly installed Starlink terminals, bypassing the normal firewalls and logs between sensitive systems and the terrestrial internet backbone…
A Scandal at the Center of the Economy
This scandal holds an irony that’s especially befitting of the AI moment. Huang seemingly has cozied up to his industry’s No. 1 foe, China, the boogeyman that, by the professed logic of Silicon Valley’s AI executives, is forcing many of his peers to build a product that they fear. Instead of denouncing him, Washington and Silicon Valley are motivated to circle the wagons around Huang, in part because doing so keeps the system moving and the money flowing. The build at all costs rationale that dictates treating China’s AI prowess as an existential threat is also driving the same people to brush past an egregious dalliance with that enemy.
This is the type of warped incentives that the AI race produces. Playing the game necessitates embracing preposterous risk to justify an idealized future outcome. To run this race requires its participants to withstand intense cognitive dissonance and behave in ways that make them appear ever more inscrutable and reckless to those who are not involved. If nothing else, Huang is a useful example to understand what happens when the fortunes of countries and many of its most powerful citizens are all tied up in one progressively intricate technological and financial bet. In this way, Huang is the AI boom personified, a hyperobject of a man who has wound his way into the center of the economy and whose influence, depending on how you look at it, may usher in either prosperity or untold doom.
That’s it, I’m trying a 9front this weekend
Post screenshot of forum from within new OS, as is traditional
Will do!
Any other no slop instances? especially that don’t federate with the proslop ones?
