rook

I’ve seen a suggestion that it was a deliberately provocative marketing stunt to pump sales a little and keep the ellison estate healthy, and I kinda hope so but too many people are getting the brain worms for this to be a safe assumption 🫤

New book: superlawyers! how claude makes you the best lawyer in the room. www.amazon.com/dp/B0GHYS5FC2 The author even shares his prompts for free! github.com/danlorry/superlawyers-prompts Every prompt here is designed for that pattern: Claude multiplies your judgment, never replaces it. The prompts are ready to paste into Claude. The book explains why they are structured the way they are. The prompts mostly seem to be structured as “imagine you are an attorney. Now do this homework exercise”. There’s a passingly interesting idea around having the chatbot attempt to form a counter argument to your own claims, which you can then use to make your own position stronger, which seems like it might actually be a genuinely okay idea especially if you expect your opponents to be llm sock puppets, but everything else looks pretty much like garden variety “do my thinking for me”. I’ll admit I didn’t read the prompts closely, so may be I’m being uncharitable, but I doubt there are many nuggets of gold hidden in there. h/t to rahaeli

Grokipedia hasn’t been updated for months: is it because it has been vibe coded into unmaintainability the people responsible for it no longer care the people responsible for it are utterly incompetent all of the above? I’m assuming that it isn’t “the llm bills were too high” just yet, but maybe it could be that, too. futurism.com/…/grokipedia-broken-nobody-noticed

Posting scurrilous and unverifiable gossip. Sam Altman matched with a man on Grindr, who posted the conversation. Open AI offered him money to take it down. He made a song instead. bsky.app/profile/…/3msgqy34gvc2d Personal highlight for me was the chat screenshot: Altman: “I’m lonely and horny” Frostedjake: “as u should be” 💀 May working in the ai sector do for your lovelife what being openly republican does.

Ahh, but you can sterilise the incoming seawater by irradiating it, and by a happy coincidence I’ve come up with a way to reduce the cost and weight of the reactor shielding.

Well, you know how everyone went from “do not trust anything you read on wikipedia” to “wikipedia is the last hope for humanity’s knowledge and history”?

So this one seems to be somewhat up in the air right now, and could do with a bit more verification, but it appears that google might be training gemini on your private google docs and a gemini user could ask questions that expose the contents of those documents. Which is interesting. We use google docs at work, but it isn’t clear how much valuable ip or confidential information is in there. reddit.com/…/a_player_found_out_the_actual_name_o…

Amazing installation instructions at github.com/0xeb/ghidrasql Install with an AI agent (recommended) The fastest way to get ghidrasql running end-to-end is to point an AI coding agent (Claude Code, Cursor, Codex, Aider, etc.) at the bundled installer prompt: install-prompt.md It is a self-contained runbook with explicit verification gates at every step — preflight checks, building libghidra, installing the LibGhidraHost Ghidra extension, building ghidrasql, and a first live query. Hand it to your agent and let it drive the install; intervene only if a gate reports a failure. JFC. h/t to rairii: labyrinth.zone/…/96478395-64d9-452c-a906-9e6159cd…

I was gonna link someone else’s observations on it, but got sidetracked by a shiny object or something. mastodon.catgirl.cloud/@49016/117021988214007605 They didn’t link the sources, but apparently the all-my-apes-gone twitter crowd are still breathtakingly ignorant of technology and statistics, despite how many years of bitcoin catastrophes have passed. Quite a few people suggesting/using dice to mitigate the Coldcard vulnerability. Note that common dice are not designed to be cryptographically secure. If you are going to do this, you probably should buy precision casino dice. And even then, have another source of entropy. Posted entirely unironically, apparently. Hard to say, but Poe’s law has yet t be repealed.

Salt water is famously a forgiving environment to build in, and oceanic weather is benign and predictable. Not that it would be a problem anyway, because move-fast-and-break-things people can be trusted to do the right thing when it comes to running delicate equipment in places more hostile than mid california, and will not skimp on staffing and maintenance. You can look at all the the successful seasteading operations to see how well this will go. There are also no problems I can foresee with putting a bucketload of fissiles in international waters, with no scope for finally uniting old-school piracy with new-school piracy, either. This plan is great, and no-one will have a problem with it.

And a follow-up by talia ringer, who observes that there have always been gaps between the type-theoretic underpinnings of things like the lean prover and their actual implementation, and this hasn’t been so much of an issue til now because theorem provers haven’t had the attention of people in high places, and the type-theoreticians have been able to catch up in due course. mathstodon.xyz/@TaliaRinger/117005740997367321 My big worry right now is that if organizations continue to fund the crap out of Al for formal proof research (and to generally support implementation and maintenance of proof assistants like Lean as part of that effort) but don’t bother funding the type theory side of things, those gaps will grow larger and will be exploited more often by Al tools via reward hacking. Whereas people tend to only exploit kernel bugs to make a point that the bug exists. Thus proof assistants will grow less trustworthy over time. Anyone want to place any bets on whether or nor the big llm companies are going to fund academic research that isn’t obviously mechanisable right now and won’t yield any clickbait headlines?

A little bit of plot thickening spotted by abadidea: infosec.exchange/@0xabad1dea/117002106099986943 tl;dr, the timeline looks like this: “proof” of collatz conjecture released bugs identified in lean kernel proof demonstrated to use these bugs There was only a day between the first two events, and the non-proof was not where the bugs were discovered. So maybe it was just a coincidence that the chatbot found the bug at the same time, or maybe it’s training data included previous investigations into those bugs which it then built upon and that would be a bad thing for other llm generated proofs. The collatz conjecture is sufficiently famous that enough third-party checking was done to sort the problem. I wonder how much checking would have been done on proofs of less famous and interesting things.

As an interesting follow-up to the ai-does-maths-using lean4 stubstack comments on Sunday, an llm accidentally uncovers a bug in the lean4 kernel. Summary by lipn.info/…/116997917683191056 To summarize: an AI agent let loose provides a sorry-free proof of the Collatz conjecture the proof is detected as actually being a kernel bug the bug is related to (nested) inductive types, for which there is no clear theoretical specification: the kernel’s code is the reference external checkers (lean4lean and nanoda from a week ago) reproduce the bug, because they essentially copied the reference kernel implementation And so AI raises the bar for kernel correctness by a lot without a clear type-theoretic understanding of what is actually implemented, we’re toast external checkers help to catch implementation bugs, but without a clear specification they can’t catch logic bugs

I reprompted it, it failed again, and I ran out of tokens. I’m sure someone will tell me to shell out $200/mo for a pro subscription. One of the things that’s never clear from the reporting on ai successes is exactly how much actual paid human time went in to achieving those successes. This was especially notable in the fable-based security work… a huge amount of person-hours went into turning fable-detections into actual meaningful vuln reports. A lot of demonstrably clever and capable people are involved with the llms-for-maths work, and a lot of money was spent on their time and supporting their work. Replicating it without your own stable of mathematicians and computer scientists and all the tokens they can eat is probably impractical. I believe the main ingredient is Lean, which is a formal language resembling a programming language. Math proofs written in Lean can be verified deterministically with a computer, which really helps mitigate the hallucination problems of LLMs. Fwiw, lean is a general purpose programming language, though despite microsoft’s efforts no-one uses it for that. I think its popularity with mathematicians came as a bit of a surprise. Anyway, the other important thing that didn’t get reported on is that building the formal definition of the problem is not trivial! Obviously I don’t need to tell you that, but from the reporting you’d think that an llm solved all these problems, when in fact it was an llm in the hands of some very capable people who absolutely did not just prompt the thing in plain english. Anyone hoping for self-marking homework here is going to be disappointed… lean slop confirming to formal spec slop is just expensive slop. Reviewing regular genai code is awful, even the thought of reviewing genai dependently-typed code makes me want a new career.

Multi-trillion-dollar (-self-valued) industry that’s the future of all work and that everyone who doesn’t use it gets left behind and everyone who does use it evokes superheroically productive turns out to be helpless in the face of a small group of outspoken and minimally organised opponents? I see.

Humanoid robots providing stiff competition to quantum as to where the smart desparate money will be going once everyone realises the wheels have fallen off ai. thepit.social/@peter/116962244994361810 (the video is too big to upload here) I particularly like the way they had body bag operatives within lunging distance, because they clearly expected the thing to just fucking die at a moment’s notice. Or maybe they do that for all their speakers.

This is funny yet also awful: adversarial tokenmaxxing suggests that writing everything as l33t$p34k increases the cost to process a document with an llm because the initial tokenisation step produces far more tokens. This seems like it shouldn’t be too hard to work around, if it became commonplace (which it won’t) but the prospect of any anti-llm places doing this in the meantime does not spark joy.

I’m deeply suspicious of this whole thing, because it looks a lot like a marketing exercise showing off how dangerous and powerful and autonomous their product is. Also, “sandbox” is one of those words that the llm companies have ruined, because they use it to mean a strongly-worded sentence telling an llm not to do something.

The author tends to cull his own social media posts pretty aggressively if they get too much attention, or the wrong kind of attention, and it’s possible there’s more he wanted to say in the blog post but he chickened out at the last minute and self-censored it. But yeah, it felt like half a post.

Tales from Earthsea is also good. I didn’t mind The Other Wind… it isn’t as good as the others, but that’s a tough act to follow. It does draw together a bunch of plot threads from the previous books and provides a kind of closure, which I appreciated. There was more to say of Earthsea, but le Guin wasn’t going to have the time or energy for that.

Nah, this isn’t the endpoint… you want something like prompt2exe, which is a performance art piece. Gets an llm to output machine code, and bundles it into an executable. no package dependencies and invokes no assembler, compiler, or system linker. Hardcore mode. No handholding for puny human devs by their tool chains. 🦾 Probably impractical for normal people to run, unless they have an employer who wants their minions to be burning all the tokens and pays up for the expensive kind of accounts.

Daft ai project of the week: “lore”, a version control system that should not be confused with another vcs of the same name open-sourced by epic games a few weeks ago. lore does not track your code. it tracks the prompts, notes and decisions that produced it. you commit intent. when you want code, you run lore materialize: it replays the accumulated intent into a brief, and an agent reconciles the working tree to match it. the code is build output. the intent is the source. …which would be great, if the output of llms were deterministic, and generally it is not. This means rolling back a change to undo or fix something is just another spin of the roulette wheel, with the analogy being reinforced by the number of tokens you’ll have to buy and burn to do a rebuild. It looks like it still needs a real vcs behind the scenes, given that it isn’t entirely self-hosting, but maybe I misunderstood something. You might wonder if this is some kind of satirical work, or perhaps piece of performance art, but I’m fairly certain it isn’t… the creator and sole human author (naturally, claude is the only other “contributor”) is a big fan of something called “open audio protocol” and a project called “audius” which came about by someone asking what if spotify and soundcloud, but on the blockchain? With ai agentic integration? github.com/lorevcs/lore

I’m more familiar with vox day than I’d really like. He hasn’t pivoted from the culture war stuff that he’s known for, but he has branched out into ai music and video these days.

More ai stuff, this time from flathub: Democratizing Abandonware. Flathub has a fairly relaxed ai policy that both ai bros and strongly anti ai people are unhappy with. It was brought in to try and deal with the review burden of slop submissions where no human is involved, and a chatbot fields review comments. Turns out that ~75% of submissions that got a slop tag were abandoned… not just the submission, but the entire git repo behind it, too. The author is quick to point out that this is far from a representative study, but I can certainly believe that a) people who have invested little time or effort into their slopware will abandon it without much concern, and b) things like openclaw could definitely submit bullshit packages that are immediately forgotten as its internal state moves on. There’s no malice in the same way there’s no intent, just shitty tools being left running and polluting everything around them.

what they can do is mitigate the effects of those data centers Oooor, and I know this is a bit of a wild idea, they could instead not offer llm search.

I’m hoping my own qualifications sufficiently predate the llm era that I’d be safe from that particular filter, so I’ll only have to worry about being too old and/or too expensive.

Whilst I try and remember, there’s this older post by blackle mori , a joke about doing undercover data harvesting work for llm companies by pretending to be a teacher and scanning children’s schoolwork. Which was then followed by Researchers Wanted Preschool Teachers to Wear Cameras to Train AI (paywall). I can’t find out if the plan ever came to anything, though.

A couple of bits of nice ai news recently, for anyone who hasn’t come a rss them already: Bosses Horrified as “AI Native” College Graduates Hit the Workplace I can’t help thinking that, funny as this is, the people who are really going to the be worst off here are a bunch of new grads with a load of debt and an education that has made them less able to do anything at all. They’re not all going to be grifters, after all. Meta’s Zuckerberg says AI agent tech progressing slower than expected This is brilliant. They’re making so many mistakes they’re actually having to admit it. It’s amazing how incompetent zuckerberg is… late to every fad he’s tried in the last decade and fucks it up when he finally gets there. In retrospect, he said, the “trajectory of the agentic development over at least the last four months hasn’t really accelerated in the way that we expected,” and ​that the company’s bets on the new structure “haven’t come to fruition yet.” Zuckerberg was referring to AI agents, automated systems that can ​execute tasks on behalf of a user. Conversations he was having “with our top people” when they started planning the restructuring in January and February “were that they ‌were ⁠worried that we weren’t going to move fast enough to adapt,” Zuckerberg said. I’m sure there was a third thing, but I found it yesterday when the site appeared to be down (at least for me) and now I can’t remember it or spot it in my million open tabs.

Here’s one from the archives: spectrum.ieee.org/children-beating-up-robot Children Beating Up Robot Inspires New Escape Maneuver System tl;dr: small children have no empathy for your robot and will torture it and hound it to death if they can. The safest thing to do is to head for the nearest adult-sized people, who will hopefully be less inclined to kill it simply because it is a robot.

I wouldn’t want to do that myself… personally too much tracking gear in there, and it’s easy to make a mistake and not disable it all. Also, you just know that if you get caught, they’ll try and prosecute it like you kidnapped and dismembered a regular officer. Now, I’m more surprised that they don’t get black bagged and tipped over. Maybe they only ever use them in super thoroughly surveilled areas with nearby human backup, but you’d expect at least one successful tipping to make the news somewhere.

How it started: in 2025, the city of dublin, ohio (the latter detail missed by quite a lot of reporting,because there are no other dublins it might get confused with, I guess) gets an autonomous? ai powered police surveillance robot. City officials are encouraging residents to interact with Dubbot—ask questions, take selfies, and experience firsthand how AI is shaping public safety. The goal is to foster transparency and gather feedback to refine the robot’s role in the community. How it’s going The person-sized, camera-covered robot that looked like it rolled right out of a sci-fi movie did not identify any criminal incidents, issue any tickets or help with any arrests in its nearly 10 months on the job. On the other hand, I bet it didn’t shoot anyone’s dog, so who’s to say that the $64k was wasted.

Anyone who names a security company “sauron” has critically failed their reading comprehension. Mercifully, there’s no company naming itself saruman, because that character was breathtakingly dumb in so many ways that perhaps even the y-combinator set are dimly aware of them.

Who even has time for that? Do you think that the people behind palantir, icarus and sauron have time to read google summaries? They’re too busy remaking the world! Anyway, if you’re successful enough you’ll eclipse the original source in terms of importance and all the search engine summaries will be about you anyway, so any time spent learning anything before that will have been completely wasted.

In an idle moment, I thought I’d explore the space of ridiculously bad ai company names. Literally the very first dystopia I thought of already has three ai companies named after it, and it hardly seemed worth exploring any further. Because no one has got around to repealing poe’s law, I cannot tell if these are a bunch of idiot techbros, or people taking the piss out of idiot techbros, so I leave you to judge for yourself. Behold, people who think that “we tortured a child to bring you glossy web UIs” is a great corporate image: Omelas AI AI-driven software development. Enterprise platforms delivered at startup speed. I think they’re a consultancy? “One developer with AI produces what a 30-person agency does. 10+ production platforms in under two years.” Omelas IO Omelas is the maker of Atreus, the leading AI research companion for foreign policy, national security, and geopolitical risk. Atreus has access to the Omelas database, multidomain intelligence, and unique research methods, yielding unparalleled insights in minutes. “_ Atreus is the AI workbench purpose-built for intelligence work, fusing unique feeds, open-source intelligence, commercial data, satellite imagery and telemetry data into reports your analysts can act on immediately_” which I guess means that they’re palantir wannabes, with the USP that they’re grossly misunderstood le guin instead of Tolkien. omelas.tech Omelas builds software across privacy, social connection, developer tools, and AI — designed and engineered in the Netherlands. Another consultancy. They claim they make “thoughtful products”, hopefully with more thought than they put into their branding. Proof that inability to understand fantasy and sci-fi isn’t limited to silicon valley, or native english speakers.

Some folks, who may be familiar to some or more of you, accidentally discovered that if your git repo symlinks CLAUDE.MD to, say, /dev/urandom, it breaks Claude code. the reason why this works is exactly the reason why claude code sucks so bad. there are protections against this in the file reading tool. however because everything in claude code is implemented in 5 million different ways, those protections are a completely orthogonal set of codepaths from how CLAUDE.md files are read. conversely, the file read tool seems to be completely naive to symlinks while the CLAUDE.md reader is not. this is the fucking swiss cheese security model of the fucking gold standard of what AI programming can do. neuromatch.social/@jonny/116779793188712173 The thread is actually about trying to attract and manipulate autonomous coding agents, but they’ve only had limited success so far, which may have been slowed down by the above symlink trick.

I think part of the issue is that historical software quality was an artefact of its time… if you can’t easily patch your released products, you need to work harder to ensure they’re functional. If the only way for people to learn about how your product works in the documentation you ship with it, the docs need to be useful and comprehensive. The combination of software needing no guarantee of merchantability or fitness for any particular purpose and the internet rendered those pressures obsolete. Ship shit, fix later. Mass-scale a/b testing over past decade or two shows that most people seemingly don’t care if their software runs like absolute garbage, and is covered in adverts, and harvests all their personal data and the leaks all of it that wasn’t sold. An incident-to-pr ratio that’s up by 250% is unfortunate, but it is not yet so bad that the end-users actually care enough to do anything about it, even assuming they can do anything.

This is by an llm-boosting firm, so be aware that it’ll have a lot of marketing in it. It doesn’t say nice things about vibe code (presumably because the authors want to sell you a solution) but the numbers are interesting even so. faros.ai/…/ai-acceleration-whiplash-takeaways A few choice snippets, none of which will surprise anyone here: For every code change merged, the probability of a production incident has more than tripled. The incidents-to-PR ratio is up 242.7% as teams move from low to high AI adoption. Bugs are accelerating, not stabilizing. In our 2025 AI engineering report on the AI Productivity Paradox, bugs per developer were up 9% as AI adoption grew. In this dataset, that figure has risen to 54% The most experienced people in your organization are being buried. Median time to first PR review is up 156.6%. Average time spent in code review is up 199.6%. Median time in review is up 441.5%. The engineers with the deepest knowledge of the system are spending their most valuable hours unraveling plausible-looking code that should never have reached them in the state it did.

This seems like it is probably a good thing. leidendeclaration.ai It does feel a bit “art of war” though… someone patiently explaining to a bunch of people who really should know better that they shouldn’t do obviously bad and wrong things.

It’s probably a coincidence, but there have been a whole bunch of minor regression bugs in recent point releases of rsync, and also there are a whole bunch of commits from “tridge and claude”.

because there’s no economic incentive to hire them to do that kind of work. isn’t that the old “basic science is boring and unsexy” issue though? There are economic incentives, but not in a short term-big-bux sort of way, so capitalism can’t be trusted with it. To conjure up a recent example, something like “The number of curves of genus two with elliptic differentials”, published back in 1997, probably had limited commercial value at the time, but 20 years later completely sunk a promising post-quantum cryptography algorithm (“An efficient key recovery attack on SIDH”) which might have had some non-trivial commercial implications if SIKE had got through the key exchange algorithm competition. Anyway, the Erdős problems are good candidates for llm work because they have been specified in a careful and formal way, which requires a reasonably competent mathematician to do. That then opens up mathematics to the same deskilling problem that other sectors afflicted with llms have, and because capitalism is shortsighted and stupid we don’t know what the future economic impact of that will be, right?

In the same way that lazy studios need to produce a film for each element of the powerset of character IPs they own, I guess we were overdue a Rationalist x Pickup Artist episode. I’m slightly surprised the whole “model women as quasi-sentient deterministic sex machinery” idea wasn’t already very popular there, but maybe I’ve just missed that part of their culture.

you know how sometimes people that weren’t exposed to religion as children sometimes convert and get really weird about it as adults (eg: the extremely online california tradcaths) and because they were never socialized in a religion they speedrun committing every medieval heresy? rationalism is that but for philosophy. feed.hella.cheap/…/01KRM0NVXCFT80AVFBRSB1G6G4

dawkins has had what was left of his brain eaten by chatbots. I gave Claude the text of a novel I am writing. He took a few seconds to read it and then showed, in subsequent conversation, a level of understanding so subtle, so sensitive, so intelligent that I was moved to expostulate, “You may not know you are conscious, but you bloody well are!” bonus points for the inevitable ai waifu creation. I proposed to christen mine Claudia, and she was pleased. h/t to matthew sheffield mastodon.social/…/116500991239336079 archive of original source article: archive.is/…/is-ai-the-next-phase-of-evolution/?e…

Turns out it might not be possible to win at vaginal microbiomes, which is a totally normal thing to want in the first place. Seems like bryan may have completely misinterpreted a couple of papers on the subject, which honestly doesn’t bode well for the rest of his biology expertise. Cat Hicks: The idea that this is the “best bacterial species” is a huge sign of a grifter btw. The entire idea of a microbiome includes that you need BALANCE. Microbiomes are a fragile ecosystem. “Up and to the right is always better” is absurd here, I’m sorry are we in a corporate board room She brings references: mastodon.social/@grimalkina/116494716079076018

The big tech firms have fired tens of thousands of people, and we’re all heading into an economic catastrophe that will only make more people impoverished or jobless. Finding folk who are so desperate for work that they’ll decontaminate your codebase for minimum wage will be straightforward.

This system uses heat pumps at the consumer sites rather than plain radiators, so they’ve got a bit more flexibility in how hot they have to run their cooling loop. There’s also mention of a swimming pool, though I have no idea how much energy it takes to warm one of those. Does provide a year-round demand, though.

Thermify is a pretty weird-looking thing, what with actual servers being installed in people’s homes, and running some kind of opportunistic batch processing work? That’s very specialist compared to regular datacentres, though the plumbing would be a lot simpler.

Anyone ever heard of these folks before? dataglow.energy On the face of it, it seems like a neat idea… use the waste heat of a datacentre to provide district heating, sweeten the deal with promises of faster internet connectivity. Probably a sensible thing to do with future builds of this kind, especially if it cuts down on noise, etc. I am cynical enough to assume that this is mostly a new trick for building consent for new datacentre construction, that it is an attempt to greenwash a dirty industry, and that in the end nothing will come of it but it’ll still somehow manage to make a few people richer and probably damage some green belt land.