Want to wade into the snowy surf of the abyss? Have a sneer percolating in your system but not enough time/energy to make a whole post about it? Go forth and be mid.

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 so 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.)

Starting this Stubsack off, I've found another FOSS project that hit the digital krokodil - ntfy.sh v2.18.0 was written by AI

Never have so few been so unsatisfied to be so correct.

Missed this when it was first going around

Leave it to fucking zuckco to figure out the worst way to make They Live happen

I feel like at this point I want to highlight the ones that took a clear stance against LLM code. On a chardet thread, people listed:

• Gentoo
• Servo
• Loupe
• Qemu
• postmarketOS
• GoTo Social
• Zig

guess i'll have to write my own unifiedpush provider

I’m still happy with Pushover. Hasn’t changed in a decade (and a half?! Been using that since 2012, damn), works really pretty well.

It’s not self-hosted but when there are push notification services on the path, nothing really is.

there's also overpush, which is meant as a self-hostable drop-in replacement for pushover and does not use ai afaict.

Anthropic is suing the Pentagon

This whole saga is a resounding “everyone sucks here”. but I’m gonna have to side with Anthropic on this one because at least they have some incredibly basic standards, which is far more than I can say for the current government and OpenAI, though the real best outcome is if the government and the AI industry destroy each other

(this has now been deemed high-quality enough for its own post)

Does anyone know a summary of the shakeup at CFAR in 2016? In January AnnaSalomon promised LessWrong that "CFAR's mission is to improve the sanity/thinking skill of those who are most likely to actually usefully impact the world." In December she announced a pivot to preventing the Reign of Steel. Julia Galef left that year and has not been very visible since. Her husband Luke Muehlhauser is OpenPhil's Managing Director for AI Governance & Policy so still Roko-curious.

LessWrongers sometimes say that Michael Vassar influenced the curriculum of CFAR's workshops even though he was no longer employed by a Rationalist charity. Brent Dill was living in Berkeley participating in rationalist events at that time.

Revealed: UK’s multibillion AI drive is built on ‘phantom investments’

https://www.theguardian.com/technology/2026/mar/09/revealed-uks-multibillion-ai-drive-is-built-on-phantom-investments

Scratch a Moldbug and find a neocon

Really weird focus on Brooklyn.

to what extent does he actually believe this? is that even a meaningful question? i think this narrative is way too esoteric and absurd to really convince anyone, so it doesn't even appear valuable if his goal is to flood the zone with post-truth nonsense.

whoever loses, we win

I somehow missed how American leftists were instrumental in urging the Iranian people to oust the Shah. Leftists like... Jimmy Carter[1]

edit this is typical US-centrism, other people don't have any agency, it's all about America

[1] I know, I know, about as left-wing as Genghis Khan

The specific article's framing pisses me off...

Anthropic CEO Dario Amodei picked a major fight with the Department of Defense last month, asserting that his company’s AI models couldn’t be used for mass surveillance of Americans or direct autonomous weapons systems.

As to who picked a fight with who, the DoD wanted to change the terms of their contract, to which Anthropic apparently compromised on every term except for mass surveillance of Americans (fuck the rest of the world I guess) and fully autonomous weapons (cause a human clicking "yes to confirm" makes slop-bot powered drones so much better). This wasn't good enough for this authoritarian strongman administration, so Pete Hegseth took the fight public with tweets first. So the article framing it as Anthropic "picking a fight" is a bullshit framing. I mean, they did kind of bring it on themselves hyping up their slop machine like it was a sci-fi AGI, but they didn't start the fight.

For one, “it’s 100 percent in the government’s prerogative to set the parameters of a contract,” Snell & Winter partner Brett Johnson told Wired, effectively meaning there may be very little chance of an appeal.

So they find a quote about contracts, but a Supply Chain Risk isn't just the DoD deciding on contracts, it is a specific power that has specific mechanisms set by legislation. If (and it is a big if with the current Supreme Court's composition) the court actually considers the terms set out in the legislation (including, most problematically for the DoD, a risk assessment and consideration of less intrusive alternatives), I think the DoD loses. Of course, the SC has all too often been willing to simply defer to the executive branch's judgement, even if the process for the judgement was "Trump or one of his underlings made a choice on a spiteful or idiotic whim, announced it on twitter, and the departments underneath them rushed to retroactively invent a saner rationalization". If the DoD decided to just end the contract (without all the public threats of SCR or invoking the Defense Production Act) Anthropic wouldn't be in a position to sue and this drama wouldn't have been as publicized in the first place.

But the lawsuit itself takes a dramatically different tone.

Yeah because one set of a language is a CEO trying to grovel and backtrack on one of the rare few ethical commitments he has ever made, and the other is making a court case about the actual law.

If the DoD accidentally pop the AI bubble by triggering a cascade when Anthropic runs into issues; then later the DoD loses the court case in a humiliating enough way; then DoD loses a civil case with the money going to pay the debts owed in Anthropic's bankruptcy proceedings, and the American public blames all of (without letting one shift the blame to the other) the Trump administration, the Republican party, the parts of the Democrat that acted as pathetic enablers, and the tech ceos for the following economic depression... I would count that as a relative win?

The four-day live-in rationality workshops at CFAR remind me of the live-in blog fests and conferences at Lighthaven. Someone in the comments to the January 2016 posts asks why pay $4,000 for a workshop in the SF Bay Area when you can learn similar content at a college where you live or from free online courses (the commenter later recanted this blatant heresy). Its hard to argue that in-person events in the SF Bay Area are an efficient use of funds, but they let people who already live there keep themselves busy.

Hello from the Center for Applied Rationality! ... We have a new experimental mini-workshop coming up soon (June 2025) and hopefully more workshop content to follow after! ... Pricing is $750 for the CFAR event, plus another $450 to sign up for Arbor (at Lighthaven in Berkeley). This is notably cheaper than the $3900 we've historically charged for most mainline CFAR workshops, since it's a more experimental program -- future workshops will likely be more expensive than this test. https://less-wrong.livejournal.com/4396115.html

This post claims that they could not find anyone doing anything similar https://acritch.com/cfar-scaling/ I know a US military veteran who had a critical thinking course which he pulled out whenever he had a training day to occupy, so maybe they needed to look outside their bubble?

Hmm, he's still sticking to tweet-threads on Twitter. We'll know he's fully cracking when he resorts to Ackman-style unreadable text blocks on there.

"Aging left" has lost "vitality" - he's phoning this one in, straight out of the house style guide.

Previously, on Awful, I predicted that Oracle would be all-in on the bubble:

Microsoft knows that there’s no money to be made here, and is eager to see how expensive that lesson will be for Oracle; Oracle is fairly new to the business of running a public cloud and likely thinks they can offer a better platform than Azure, especially when fueled by delicious Arabian oil-fund money.

But, uh, there's not going to be any Arabian money while we're dancing in the desert, blowing up the sunshine. The lawnmower is now running low on gas. Today, Oracle continues to make astoundingly bad business decisions:

Oracle is the only major player funding the AI buildout with debt, carrying over $100 billion on its books while free cash flow has gone negative.

https://helenofdestroy.substack.com/p/grand-theft-reality h/t naked capitalism

Those interested in upgrading to the full RealityPlus™ experience will soon have not one but three styles of brain chip to choose from, expanding Big Parasite’s vertically-integrated propaganda pipeline into a perfect server-to-cerebrum delivery system while realizing the transhumanist dream of merging with the machines. Sam Altman’s brain-chip company is even called Merge Labs, because subtlety is for poor people. Yes, the guy who says human children waste more energy than OpenAI’s planet-liquidating data centers will be playing tug-of-war for direct access to your cognition with Musk and Mark Zuckerberg. Coverage of this assault on privacy already reads like articles about AI from five years ago: You don’t want a brain implant? Are you some kind of Luddite? Better get over it: “avoiding brain-to-text devices will feel like avoiding smartphones.” It’s not like Meta’s underpaying African contractors to watch you through your augmented-reality Raybans while you shit or something. Why is Meta’s glasses project head Rocco Basilico seemingly named after Roko’s Basilisk, the AI bogeyman who will go back in time to torture you if you don’t help create it? Is Roko’s Basilisk…Jewish? Remember to smile for Sam Altman’s soul-sucking WorldCoin orb or you won’t get your UBI!

CFAR seems to have pivoted back to focusing on the workshops. Their winter 2025/2026 fundraiser only raised $10k with a goal of $125k. The curriculum sounds very New Age:

If you’ve been to a CFAR workshop in the ~2015-2020 era, you should expect that current ones: ... Have roughly 1/3rd new content, mostly aimed at practical ways to be less “seeing like a state” when applying rationality techniques, and to be more “a proud gardener of the living processes inside you / a free person with increasing powers of authorship.” (We've been calling this thread "honoring who-ness.")

No masks in their photo of a workshop posted February 2025 (2024 was a pretty bad year for airborne infections where I live, and alienated educated young people are more likely to wear respirators than normies, so I would expect to see someone in that room wearing a N95 or Flo). If building warm and nurturing relationships is important then it helps to be able to eat together and see each other's faces. The venue is about a 90 minute drive from Oakland, CA (the East Bay).

This paragraph leapt out at me:

On Day 4 of the four day workshop, we spent three and a half hours on an activity called Questing, in which participants took turns being the “hero” (who worked on whatever they liked) and the “sidekick” (who assisted at the hero’s direction) for ~10 minute chunks. This activity was extremely well-liked (did best of all activities on our survey; many said many great things about it).

If you read that and say "doesn't that sound like Effort Exchange in the Dragon Army Barracks?" you should go home and rethink the regrettable things you learn on the Internet. I look forward to reading the book on LessWrong, the splinter sects, and just how much they had in common after a hard day gardening in a post-apocalyptic wasteland.

Before FTX collapsed my model of LW was something like cryptozoology enthusiasts who trade posts and sometimes meet at a con, now its more like Scientology. Early Scientology offered a community and a path to self-improvement.

Tudeh are western stooges

Moldbug 🤝 Iranian Government

Ooh, that’s really cool. Also, very sobering section on the various e2e methods that are a pretty thorough indictment of all the chat systems out there.

FT reports from Amazon insiders that they're investigating the role AI-assisted development has played in a spate of recent issues across both the store and AWS.

FT also links to several previous stories they've reported on related issues, and I haven't had the time to breach the paywalls to read further, but the line that caught my eye was this:

The FT previously reported multiple Amazon engineers said their business units had to deal with a higher number of “Sev2s” — incidents requiring a rapid response to avoid product outages — each day as a result of job cuts.

To be honest, this is why I'm skeptical of the argument that the AI-linked job losses are a complete fabrication. Not because the systems are actually there to directly replace the lost workers, but because the decision-makers at these companies seem to legitimately believe that these new AI tools will let their remaining workforce cover any gaps left by the layoffs they wanted to do anyways. It sounds like Amazon is starting to feel the inverse relationship between efficiency and stability, and I expect it's only a matter of time before the wider economy starts to feel it too. Whether the owning class recognizes what's happening is, of course, a different story.

So oil prices are down again, and on nothing but a promise from Trump and a promise from the EU. The economy has proved remarkably resilient to me; the attack on Iran is like, wild nonsense number 17 that the USA regime did that I thought would trigger a major recession, and didn't.

I mean don't get me wrong, things are much worse now than 3 years ago, clearly. But they're not like, Great Depression worse. They're not even 2008 worse. It's just a certain level of degradation (cost of living is higher, purchasing power is lower, concentration of wealth is higher etc.) that people got used to as the new normal. People can get used to lots of things.

To make the IT analogy, I think the global economy is like Twitter. Sure, it feels like a Jenga tower held up by thoughts and prayers, but it's holding up. When Musk took over I really did think his catastrophic management philosophy would completely break Twitter, but no, it trudges on. Yes, moderation is now nonexistent, and I'm told it's down more often, and often in "soft downtime" like notifications not working, or DMs, or some other feature, or it's working but slow, and so on. But clearly the site is up most of the time and more or less functional. Users just get used to degraded quality as the new normal.

I predict AWS will 1) get slower and costlier thanks to "AI", with higher downtime, at higher stress for the workers; 2) the leadership will refuse to see or admit or even consciously be aware of this; 3) the worsened services will be the new normal. I predict similar developments for the socioeconomic situation of the world, too; though I'm not ruling out a spiral into complete recession, either.

I mean it's not too far off from the standard color revolution conspiracy theories where nefarious American intelligence agents and NGOs are working towards regime change and civil strife across the world in order to advance their sinister ideology. But where the "classical" color revolution conspiracy serves to undermine anticommunist movements in Eastern Europe surrounding the fall of the Soviet Union by positioning them as patsies or victims of the CIA, this newer variant that Moldbug is working with is trying to discredit American domestic anti-imperial/anticolonial/antifascist sentiments by positioning them as puppeteers of oppressive foreign regimes. Kind of an uno reverse card being played on the original story, but one that fits with how the American right conceptualizes itself and its domestic opposition.

Somehow I had never found that dragon army retrospective before and had the fascinating experience of wanting to explain to someone that "no, what you're describing is actually a cult. Like, you're describing being a cult leader." Which is usually not the person to whom the cult dynamic needs to be identified and explained.

to follow this one up: there is now a new study about AI agents being dogshit at keeping code working over the long term

Unfortunately the paper structure screams "AI senpai, notice me!"

AI coding agents seem bad at this job yet, but if you optimize for our benchmark...

NeoReaction: It's Always Antisemitism

That didn't really surprise me tbh, I follow a blog that is ostensibly meant to be a furry blog but gets repeatedly sidetracked into cryptographic stuff, and the conclusion for anything e2ee is essentially that only signal is worth using if you are looking for actual e2ee. But yeah, encryption is generally pretty bad on this kinda thing.

If Duncan Sabien or Eliezer Yudkowsky admitted what they were doing, they would have to take more responsibilities. If they moved from chanting "we have noticed the skulls" to expecting every serious member to be able to describe times that nerdy subcultures they were not part of went wrong, they would have to give up most of what they do.

Back in 2019, Ben Pace of Lightcone said that CFAR and Lightcone were one legal entity, but two boards with no overlap. Did CFAR + Lightcone really spend $22 million on real estate in Berkeley without spending a few grand to create a separate nonprofit and separate the finances? In 2024, CFAR still had the real estate and the mortgage on its books. https://www.lesswrong.com/posts/eR7Su77N2nK3e5YRZ/the-lesswrong-team-is-now-lightcone-infrastructure-come-work-3

I have never opened a US business bank account, but I would think it would be hard to keep the bank accounts separate if one organization has no independent legal existence, and transactions in the millions or tens of millions tempt the most righteous person to stick his fingers in the till.

It's theoretically possible to keep them separate, but I would assume in this case that it's evidence that regardless of intentions CFAR and lightcone are sufficiently closely linked to be basically the same organization. I mean, if there's not a separate legal entity then I would assume anything involving money is going to require the same person or persons to sign off on the transaction, regardless of what the board looks like.

Forming a single legal entity would have made it hard to protect the other projects if the CFAR side had lost a lawsuit over abuse of a minor at a CFAR event, or Lightcone had lost a judgment over taking money from FTX and had to sell the Rose Garden property, I know these people don't do "fear of frequent consequences of ordinary human weaknesses" but that is a big risk.

I also wonder who served as treasurer and bookkeeper for each project. If one person served both projects, he or she could have caused all kinds of trouble, even if there were separate bank accounts.

I somewhat agree although when the "other shoe drops" and these things start impacting the money men they may start to realise AI isn't the magic cure they thought it was (he says kind of hopefully)

6 hours of downtime for Amazon shopping. A very simple back of a napkin calculation. They made $213.4bn in sales in q4 2025. So divide that by 90 days and then 24 hours and multiply by 6... We are talking a $0.26bn loss for 6 hours downtime... That is not an insignificant amount of money. I imagine most bosses would be screaming for heads having lost that much money in sane non-hyper-scaled businesses.

Man, I wish I had enough money to fuck around with nonprofit shenanigans

Really not liking the ratio of tech industry sneer to ZOG NWO thought control rant in this one, honestly. (Not to mention the latter gives way too much credence to the absolute nonsense brain chippers love to spout about neuroscience.)

Yeah, soatok has been doing really good work there. It’s disappointing that the matrix folks don’t seem to take it seriously.

Silicon Valley is buzzing about this new idea: AI compute as compensation

These people are genuinely unhinged.

As the recent harpers article says:

"...people who should be in The Hague are giving [startups] twenty million dollars. Something bad is gonna happen here, something really fucking bad is gonna happen...”

"Selling your soul to the company store is not just fun, it is also invigorating!"

It's also a trend that I don't see stopping without a major structural change. I don't think there's a point at which they're going to say "we've cut enough corners and are going to stop risking stability and service degradation." The principal structure driving the economy, especially in the tech sector, is organized around looking for new corners to cut and insulating the people who make those choices from accountability for their actual consequences.

Man, that harper piece is a full DnD alignment chart of the most online bay area weirdos you've ever seen.

Has anyone heard of the Internal Family Systems Model? One of the CFAR founders said he relied on it when he was designing self-help workshops. The IFS encourages you to see yourself as a system of entities and talk to them separately, and that reminds me of Ziz Lasota's two-hemispheres theory and Michael Vassar's jailbreaking.

found a paper that may be of interest and review interest to some here

linked via this toot that hit my feed

I’ve heard of it, including in some outlets that (at the distance I am to it) seemed to pass the sniff test

but I’ve also seen it kick around TPOT

so I’d definitely want to seek out the advice of an expert if I cared about it

I heard somewhere that "there is no unitary self" can be a Buddhist teaching and TPOT draws on Western Buddhism. There is work to be done figuring out where they got their eclectic mix of techniques and terminology.

Julia Angwin:

I'm suing Grammarly over its paid AI feature that presented editing suggestions as if they came from me - and many other writers and journalists - without consent.

State law requires consent before someone's name can be used for commercial purposes.