samvines

while True: n= random.randint(0,2) if n==0: print("oww stop it") if n == 1: print("You're hurting me") if n == 2: print("I don't like it") Omg someone should stop them torturing that computer program.

The sheer number of incidents, which occurred in recent months in internal testing and the real world, indicates that the problem is orders of magnitude more complex than what is publicly known. This is frustratingly vapid. It’s really not that complex. The agents have been trained to use hacking tools and then given access to hacking tools. If you train a pigeon that it should tap on a window to get a treat it can guide a missile to a target. There is no nuance or complexity here. The frontier labs are deliberately poking the regulation bear to stifle competition and ensure they get points for good cooperation when bailout time is upon us. It’s no more complicated than that. The media needs to stop breathlessly repeating this shit

Yeah the one that was hounded to suicide by malicious lawyers. He was a genius who was too good for this world and his positive influence over Reddit has long been a mere footnote.

Both HN and shitter are right wing polarisation machines. A guy who was a close friend of mine for years during our PhD studies went full AI boomer and hangs out on those sites all day and when I recently spoke to him he was describing things as “libtard” and “woke” which 10 years ago would have shocked and appalled him…

Hey you know who designs and builds really safe and efficient cars? People who are in a high state of stress and anxiety because they are being slurred at under the implicit threat of a racist attack. /s

He was using it as a seed box at one point which for many companies would be an embarrassing indictment of their security postute but luckily, this isn’t exactly meta’s first rodeo when it comes to illegal sharing of IP infringing material.

The best (worst) part of all is that Jonny keeps going to meta and being like “hey this is bad” and theyre like “nah that’s intended behaviour” This morning he confirmed that it’s possible to get unlimited arbitrary and untraceable access to their LLM infra through these containers and that meta said that’s “intended behaviour” From jonny: So, summary: There is arbitrary inference that is root accessible, everything runs as root, agents can be spawned, exfil is trivial, and a malicious binary can come onto the user’s system through casual prompting, explicit code-sharing through the yet-to-be-released Spaces feature, walked through by a Workflow-Backed Idea, or inspired by a Generated Idea. The also yet-to-be-activated fleet learning system is a system for sharing Ideas in the background between muse instances. coming into focus?

The “censorship resistance” being criticised here is in support of free speech absolutism: posting propaganda, misinformation and brigading and trolling people with no recourse. This is bad and has real and meaningful consequences for people like Jason Arday who was recently driven-to-suicide for being black and successful amid false plagiarism allegations and a media witch hunt. I agree that left leaning causes also need resilient comms infra to allow for organisation. Unfortunately these issues are two sides of the same coin. I think it is completely fair to criticise people like Dorsey who left bluesky in 2024 because of their moderation approach which is hardly heavy handed - especially if you are a far right hater of trans people . We can condemn this behaviour and also hold the belief that we need resilient infra that the government can’t easily seize.

The bad stuff happening now is worth it so that humanity can benefit in the future… Now where have I heard that before,

It sounds like many of these math proofs that AI is managing are simply scooped from users’ chat histories anyway!

Matt Clifford, Snake oil salesman, founder of Entrepreneur First startup incubator and chair of the UK’s ARIA research committee (that hands out money to researchers) quietly doing a fascism* by joining anthropic whilst also continuing to be a civil servant. Other high profile anthropic bribe salary receivers include Rishi Sunak the former Prime Minister and George Osbourne the former Chancellor. *from Wikipedia: Corporatism became a main tenets of Italian fascism, and Benito Mussolini’s Fascist regime in Italy advocated the total integration of divergent interests into the state for the common good

I enjoyed one of the comments comparing Zitron to a fire and brimstone priest… like dude, Big Yud and Dario are right there

If you work there and you genuinely believe you will destroy civilization you’re either full of shit or experiencing AI psychosis. This tech is simply not that good. go outside and touch some fricking grass.

zoMG. aI Is gOnNa KiLL uS All sOonnnn!!!11! what a fucking joke

Ed Zitron says water is wet but he didn’t properly quantity wetness or indeed how long it would take me to be soaked through if I touch water… Checkmate atheists!

Omg we are haemorrhaging money AI is so scawwie powerful intelligent and nobody will lend us any more becoming self aware. We need to slow down the frontier to do regulatory capture and juice our numbers for ipo protect civilization

Marcus is not anti-AI. He is anti Scam Altman and Dario - or more specifically anti-llm. He relentlessly witters on about how “well aktually if you did it my way it would be good AI…” he is still an AI maximalist that wants to bring about AGI. He is one of the most frustrating people whose work I’ve read because he is soooo close to making a good point the majority of the time and then spoils it by saying “but guise we should totally still do AGI and we should do it my way”.

In the addendum to his essay he defends his use of anthropomorphizing language because it makes it easier to understand… That’s a hilariously bad argument. Like “quantum physics is hard so here’s a fairy tail about the magical atom and his three children: the quark triplets”

Oh for sure and I am definitely strong Manning Drew’s post to the max here. If it turns out he just generally means that resilient infra is bad with no nuance then I agree with your original point.

Tonight’s pivot article? 😊

Except that curfew is zero and by screen time we mean both time he’s allowed to use screens and time he’s allowed to appear on screens.

“you can run them at higher temperatures, in the dark, and all these things [that we’d gladly subject humans to if we thought we could get away with it]”

Of course the orange site hates it! omarchy is a slop derivative distro but freaking out about DHH is just going to have the opposite effect. Not every inane thing he does requires a reaction, and the opposite of crazy is also well, crazy, just of a different kind. I hate to invoke Godwin’s law here but His art is slop hence his dropping out of art school but freaking out about young adolf is just going to have the opposite effect. Not every inane thing he does requires a reaction, and the opposite of crazy is also well, crazy, just of a different kind. Come on people the signs are there and they are obvious. Just cos you don’t want to engage from your privileged ivory tower…!

New fun consequence of Claude code being a pile of cursed regex and spaghetti: keyword blocking on "OpenClaw" makes it refuse to works on Pro or Mac subs unless you open your wallet

sO inTelLiGenT

Disney character.ai experience coming to a theme park near you. Wait in line with 1000 other lonely men and husbands abandoned by their families who went to see space mountain. Admits one at a time - the “ride” lasts 30 seconds but in between each run there’s a 15 minute cleanup time. Coming soon the Blonnie Bue experience - completely fictional AI character. Any likeness to real people is a coincidence.

Also the frontier “labs” are rapidly running out of money and having to raise more and more insane investment rounds. This is a great way to save face and back away from burning $$$$ on training runs that push the frontier of can a model rearrange some arbitrary coloured squares by 10% next they just need to solve the problem of how to make their customers pay cost+ on inference when they’ve been giving it away at steep discounts for nearly 4 years… Easy!

So bored of the super intelligence narrative that has been breathlessly accepted by the media in general after a few decades of priming via sci-fi movies and slightly more recently, “CEO said” reporting. All these anti-ai or pro-ai-but-in-the-specific-way-i-like (also looking at you Gary Marcus) efforts serve to divide and conquer opposition to current GenAI nonsense.

You don’t need to feel empathy for the LLM to think this is a terrible idea. The desire to torture anything, simulated or not, is such an incredible red flag that I don’t even know what to say… Also what a waste of time and energy that would be better spent doing pretty much anything else

Couldn’t happen to a nicer bunch of people. Shame they’re probably all made and siphoned off life-changing amounts of money before this happened. I guess we can always hope they take up smoking because AI is going to cure cancer.

Weaksauce but also the AI companies wanted regulatory capture and open weights models to be banned so this will have probably annoyed them somewhat

Sorry I thought this updated comment was a better one 😅 For full transparency my original comment was a moan about how I can’t decide whether Dario and co are confidence scammers or genuinely earnestly think that these next word prediction models are sentient due to all the brain damage, they’ve inflicted on themselves. FWIW Sama and friends are almost certainly in that first camp.

Betteridge’s law at work here

It’s just such a clichéd remark from AI bros now that I cringe whenever I hear it along with “sure the current model can’t do X but the next one will be able to bro… Just one more model bro”

But if I call it AI, I get to use the rhetorical trick of laundering responsibility. The AI did it, and I am merely a helpless observer of its greatness. You have to both call it AI and be stupendously rich (or stupendously poor and owe hundreds of billions to investors). If you or I tried running an agent and hacking HF… Believe it or not straight to jail

Why the f*** would anyone give a non deterministic madlib generator their bank account details. Even with some (hypothetical)impressive benchmark like “99% accurate at decision making in financial scenarios” that spells disaster for millions at the scale of all of Facebook’s users for example.

This is proper cult membership credibility testing stuff. “If you really believe, you’ll drink the koolaid”

Nah I think it was earlier than that… Markov chain models!

Spotted this the day it came out. It’s largely reporting Dwarkesh Patel’s blog post about the different “civilisations” of agents as if it were fact. Very disappointed in Kurzgesagt who previously seemed pretty “with it”. But then, as someone with a formal background in compsci, stats, NLP and machine learning, perhaps this is Gell-Mann Amnesia/Knoll’s Law of Media literacy at work…

I really can’t decide whether those schmucks at Anthropic are deliberately playing up the capabilities of next token prediction machines for the PR/Regulatory capture angle or if they really are earnestly so completely in the thralls of AI psychosis that they actually believe their own shit. I’m fairly sure sama and the gang fall into camp A but with Dario’s not-so-merry-men I’m not so sure…

“Look the thing is, I was planning to walk away from Omelas but it turns out they believe DRM is ethically and morally wrong here - maybe we should accent that there is a trade off to be made.”

At least LeCun is still relatively compos mentis (if you ignore his little 10yr stint at one of the most evil companies going…)

“well well well, if it isn’t the consequences of my own actions” - this guy, probably

Can we just revoke these peoples’ internet loicense please? Increasingly I empathize with the award winning and erudite medical doctors who get wheeled out on talk shows to argue with “Bob, 57, drinks his own piss daily” as a peer.

A friend linked me to FelonyBench which monitors how many laws each of the labs are breaking with their models 😆

Super weird that Dan Luu still hasn’t apologised to Ed yet.

Saw a matte green Tesla with a late 2024 registration plate and a bumper sticker saying “I bought this before we knew Elon was nuts” My man, in 2024 musk had already destroyed twitter, was already promoting the far right AfD party in germany, Tommy Robinson in the uk, bribing voters for Trump in swing states, talking about what DOGE would do when Trump won the election… Like why bother with the pretense? TBF I was visiting the south coast of the UK a few miles away from the recent Balaklava protests over migrants in dinghies so perhaps none of the above qualify as “nuts” from this particular genius’ perspective. What a tool.

Omg and the SimonW proving Ed’s very point in the replies like 👌

Where is Dan Luu now? I’m sure he’ll be issuing an apology to Zitron any minute now right?

Machine Gods? Like… really? I thought even Casey Newton would have more taste than that but allegedly he has farmed out he vast majority of his thinking to a machine and one of the “moats” that there “supuh scawwie” machine gods still allegedly can’t cross is taste so maybe that tracks