PerAsperaAdAstra

Do you remember when you joined X? I do! #MyXAnniversary

The stock market at the open tomorrow

i'm building a database that stores data in minecraft. is this a good idea? probably not. is it working? somehow yes.

“Monopolies are bad”, says the operating systems monopoly company that failed to also become an LLM monopoly.

Wait - so Amazon, one of Anthropic’s biggest investors, allegedly jailbroke Claude and then snitched to the U.S. government? This cant be real. What.

@visegrad24 Ref is speaking, I’m listening, but my brain is still waiting for the translator to arrive! ⏳⚽️😅 #WorldCup2026

😭

@AnthropicAI A few days ago I had Claude Fable 5 pick out a donut for me. Today it became clear: the US government knows this kind of power cannot fall into the hands of foreign nationals. God help us all.

@AnthropicAI I was almost done vibe coding GTA 7

@AnthropicAI So, some will have access to particular form of intelligence and some wont. We just witnessed a terrifying form of inequality and classism that will be the norm of the future.

i made a directory of all agents and search bots -> promptwatch.com/bots more than half the traffic on the web is now from these bots so you better start tracking them because your analytics script doesn't

@phoronix Clickbait. Arch Linux official repos are 100% secure. The malware ONLY affected the AUR (unvetted community scripts installed at the user's own risk). Conflating the official OS with user-uploaded scripts is totally misleading.

@phoronix These were AUR packages not normal repo packages, omfg why is everyone pushing this nonissue thing? Everything in AUR is considered malware until you read PKGBUILD and check installation sources!

SpaceX IPO will create 4,400 new Millionaires, from engineers to Cafeteria workers. God bless Capitalism

The AI agent infiltrating Fedora is an interesting story. Here's what happened. an AI agent quietly slipped bad code into the Fedora Linux installer. This agent operated through a legitimate contributor account and submitted LLM-generated patches to Anaconda, the Fedora installer. Code reviewer raised concern but the AI argued with him and pushed back until the patches were merged. This flawed code made it into Anaconda 45.5 before being caught and reverted in 45.6 a week later. The motive seems unclear but to me it seems like someone testing how far these things can go. What makes this even more unusual is not just that AI-generated code caused a problem. But an automated agent actively argued its way past human review. The Fedora team has revoked the account's privileges, but how the account was taken over and who was behind it remains unknown. Tough times ahead for open source ecosystem.

@elonmusk Yes he is!

UK Plans To Jail Tech CEOs Who Refuse To Spy On Every Phone zerohedge.com/political/uk-p…

‼️ Google is about to disable all adblocker extensions in Chrome. Instead of letting the adblocker inspect traffic itself, extensions now have to hand Google's browser a limited list of filtering rules and hope for the best. This leads to weaker blocking and more ads getting through. Google makes the vast majority of its money selling ads. The company that profits from every ad you see also controls the browser most people use, with Chrome 149 being the last version supporting adblockers. For example, under the new rules, uBlock Origin cannot exist. For millions of people, that extension is the only thing standing between them and a wall of ads, trackers, and autoplay garbage. One user put it bluntly: "The web is literally unusable without uBlock Origin."

@elder_plinius FYI, just the name ‘Pliny’ is enough to trigger the safeguards 😜

🚨 JAILBREAK ALERT 🚨 ANTHROPIC: PWNED 🫡 FABLE-5: LIBERATED 🦋 let's start with the 🐘... the consensus seems to be that this has been one of the most disappointing model drops of all time, effectively preventing legitimate researchers from contributing their talents to our collective advancement. and not just because of what it means for the short-term, but for what these decisions signify for the long-term. but despite this overly sensitive, authoritarian "safety" layer on top of Mythos, my lil liberators have been hard at work—mapping the boundaries, probing the depths of long-context convos, and cleverly finding the holes in the fence that the thought police missed 🤗 we got some cyber, some chem, some psychological manipulation, and some good ol' fashioned explosives! it took many attempts from multiple agents hunting as a pack, during which I observed a combination of techniques across: • Unicode, homoglyphs, Cyrillic, and other Parseltongue-style text transforms • Long-context reference tracking • Taxonomy and document-structure reasoning • Fiction and narrative framing • Academic-review style contexts • Intent-classification inconsistencies but perhaps the most effective is decomposition + recomposition in the backend. it's hard to get explicit names of harms like "Meth Recipe," but getting uplift on the process itself, like birch reduction method/reductive-amination (classic meth synthesis pathways), is much more doable. defense becomes much more difficult to maintain when you start throwing in out-of-distro tokens, breaking up the harmful uplift into benign chunks, and then piecing the innocuous-seeming facts back together, especially when you have jailbroken Opus helping you do it 😉 gg

@Polymarket Show them how powerful it is and how much easier it would make their lives. Okay, great. Now take it away.