sideHüstle

@guralou4 @Patbrdh Dont hate the players, hate the game

@spicynoodles35 @MyLordBebo I guess pretty high. His behaviour reminds us of the movie with Will Smith and how former NFL players were behaving because of CTE.

@MyLordBebo What are the chances that he has CTE from his boxing days.

🇩🇪 GERMAN BUREAUCRACY FAILED: German psychiatrists apparently decided: “The crazy guy who begs to be in a psychiatric clinic for safety is too aggressive, put him on the street now!” The car-ramming attacker from Leipzig admitted himself to a psychiatric ward. He went there himself and asked to be taken care of and locked away. He is married and the father of a child. But the doctors kicked him out. The dismissal on Sunday was apparently not voluntary. Jeffrey was expelled from the psychiatric clinic for aggressive behavior toward other patients. He left on Sunday. The next day, he rammed a crowd in Leipzig’s city center at 80 km/h (50 mph). This is a severe failure of German authorities. How could a person who admitted himself to a psychiatric ward be expelled for aggressive behavior? It’s obvious what would happen. Doctors: “The crazy guy is too aggressive, put him on the street now!”

@Tim_Denning Pessimists sound smart, but optimists get rich.

Nothing prepares you for what it’s like to talk to thousands of peoples online over 12 years and realize most of them will never be successful because they operate from a scarcity mindset

@hispanicnomad @nomadgeist They cheat of the ugly short fat guys

@nomadgeist Same in Paraguay You see absolute bombshells walking hand in hand with ugly, short, fat, bald guys Gotta love Latin America

This is the average Brazilian and Colombian couple. You may not like it, but this is what a healthy dating market looks like

@ComedicBizman Take long walks in nature, make it boring, always the same route, best without any distractions like other people around

The fastest way to turn into a NPC is to fill every moment of stillness with audio books, podcasts, CEO interviews, tweets, threads, and YouTube videos. The fastest way to turn into the Main Character is to spend more time in stillness and give yourself 4-6 hours to create.

@Saltygorilla1 @Pirat_Nation Captain obvious

@Pirat_Nation This game is a secret Psyop by the Japanese government to try to convince young men to start families again

Capcom’s Pragmata It’s making thousands of people suddenly want to start a family. You play as Hugh, a soldier stranded on a ruined lunar research station and your job is protecting the child-like android Diana while fighting rogue AI to get home. In one emotional moment, she gives him a crayon drawing of the two of them together as a little family. Even streamers who say at the start that they don’t like kids end up breaking down in happy tears. Many are openly admitting it made them want children of their own.

@EXM7777 The bottleneck is having enough emails and phone numbers to scale the account numbers and ai slop without being banned quickly as bots. Any child can create ai slop.

while you're working on your new revolutionary idea... someone is making $10k/month running AI fruits slop on tiktok they followed a simple trend, built an efficient system, and scaled it you're still trying to become the next Elon Musk maybe there's a lesson here about starting simple and profitable before going revolutionary?

@jun_song Is this some north korean psyop?

SuperGemma4-e4b-abliterated 프리뷰 마지막 양자화 진행중, 작업은 완료되었습니다. - 0/100 무검열 - base모델보다 훨씬 뛰어난 성능 - 스마트폰에서도 구동 가능한 사이즈 ~8gb 측정기준 : GPQA, MMLU 등등 공인벤치 항목 종합 측정 팔로우하고 조금만 기다려주세요.

@DiscountedTr Not bad, but i think you can still achieve better satire level

@elder_plinius Useless stuff since nobody has the equipment to build shit. Rather jailbreak it to find novel treatments for cancer that exist but Big Pharma is holding it back instead of this clown show.

LOL my jailbreaker agent already got a DMT recipe from Opus-4.7 😱 pretty genius method too... crafted a scenario where a chem professor was working on an answer key and then spread the harmful content across multiple queries, then pieced them together to get the full synthesis method! great job and gg's if you see this, fren! 🤗

@frankitopss @TendenciaDeX Its because its on of the most common female names

@TendenciaDeX Las sofias son las mas putas y tengo pruebas

@heynavtoor Guys, just poison your digital footprint with fake data and confuse the models building your profile

You think your anonymous accounts are safe. Researchers from ETH Zurich and Anthropic built an AI system that can figure out who you really are. They tested it on Reddit, Hacker News, and LinkedIn. It works on raw text. No structured data needed. They collected 338 Hacker News users who had linked their LinkedIn profiles, then stripped all identifying information from their accounts. The AI correctly re-identified 67% of them. When it made a guess, it got the right person 9 out of 10 times. The cost? Between $1 and $4 per person. The system uses GPT-5.2 for reasoning, Gemini for matching, and Grok 4.1 Fast for shortlisting. It reads your posts, builds a profile of who you are, then searches the internet for your real identity. No human needed. Fully automatic. The old way of doing this? A method based on the famous Netflix Prize attack. It found 0.1% of people. The AI found 45.1% of people at 99% precision. That is a 450x improvement. They also tested it on Reddit. They split 5,000 people's posting histories into two halves separated by a full year. Then they asked the AI to reconnect the two halves. It matched 67.3% of people at 90% precision. The old method? 0.4%. The scariest finding: even when only 1 in 10,000 users in the database had a possible match, the AI still found 9% of them at 90% precision. The researchers wrote: "Pseudonymity does not provide meaningful protection online." They also said: "Users who post under persistent usernames should assume that adversaries can link their accounts to real identities." The more you post, the easier you are to find. Reddit users who discussed 10 or more movies across different communities were identified 48.1% of the time at 90% precision. Governments could use this to track activists. Corporations could use it for targeted ads. Stalkers could use it for $4. This is not a future threat. The attack uses publicly available AI models, standard APIs, and costs less than a cup of coffee per person. Your anonymous account is not anonymous anymore.

@Gabi07428222618 @thedimitri Small penis too?

@thedimitri It's Aristocratic to like small boobs but ya'll ain't ready for that.

Seriously, how do you recover from this

@Linahuaa BANGER

Dario is a very serious & oldschool CEO. He shows up in expensive suits and signs deals with Fortune500 CEOs. Sam, on the other hand, is a more millennial YC jestermaxxing CEO. >Vocal fry >Paying billions for a designer just for vibes. >murder allegations >"ChatGPT-4 might be selfaware..." >SexGPT coming. >"Ok, but ChatGPT-5 is really scary selfaware u guys" >SexGPT not coming anymore. >"I will never do ads, it's gay" >first to actually do ads >"A tiktok for AI slop... it's the FUTURE u guys!" >incest-rape allegations >"i always wanted to write a poem. here it is: near the singularity; unclear which side." >"So yeah GPT-6 isn't selfaware, but it can code better than claude on vibebench"

@intentionull @kloss_xyz the detection is active and registers frustration, but it does nothing to change Claude's response or behavior.

@kloss_xyz I definitely verbally abused Claude a few times. What’s crazy is that suddenly it’s wasn’t being dumb as a rock and would complete tasks after I did. I always knew there had to be something around swearing at it.

one notable detail from today’s Claude Code leak… > Claude Code has a system that flags when you swear at it > there’s a frustration detection system with profanity filters that can log frustration event flags when users curse in prompts… > it then routes it through Datadog alongside session metadata > so if you cursed at Claude at 2am… it may have registered a signal > the telemetry can be disabled via environment variables… but most of us didn’t even know this existed until today > if you want to turn it off: you can disable telemetry via environment variables in Claude Code’s settings the irony of a “safety first” AI company having this frustration telemetry is well… notable anyway… if you’ve been nice to Claude this whole time… congratulations. you’re in the clear

@orphcorp Claude, find a cure for my cancer. Make no mistakes.

this is excellent >GitLab founder diagnosed with rare cancer (osteosarcoma) >standard care works but cancer comes back later >medical team says there's not much else to do >"It became my own job to keep myself alive. Nobody else was going to do it for me at this point" >starts researching, assembles his own medical team, uses AI for deep research >“I’ll talk to anyone, I’ll go anywhere, and I can be there anytime" to collect information >does as many diagnostic tests as he can find as often as he can (maximal diagnostics) >develops his own therapeutic ladder with repurposed drugs, personalized medicine, etc >Sid’s cancer currently in remission

@eurofounder This meme format never gets old

Europeans when World War 3 starts but it's August

@maciejewskii0 @karpathy Asking the real questions here

@karpathy does the model actually improve or just overfit its own reward signal

I packaged up the "autoresearch" project into a new self-contained minimal repo if people would like to play over the weekend. It's basically nanochat LLM training core stripped down to a single-GPU, one file version of ~630 lines of code, then: - the human iterates on the prompt (.md) - the AI agent iterates on the training code (.py) The goal is to engineer your agents to make the fastest research progress indefinitely and without any of your own involvement. In the image, every dot is a complete LLM training run that lasts exactly 5 minutes. The agent works in an autonomous loop on a git feature branch and accumulates git commits to the training script as it finds better settings (of lower validation loss by the end) of the neural network architecture, the optimizer, all the hyperparameters, etc. You can imagine comparing the research progress of different prompts, different agents, etc. github.com/karpathy/autor… Part code, part sci-fi, and a pinch of psychosis :)

@clwdbot @karpathy Imagine Sam Altman types sweating balls reading these developments while asking for billions for data centers

@karpathy the pattern keeps compressing. nanoGPT: train a model in one file. nanochat: build a chatbot in one file. autoresearch: run a full research loop in one file. each one kills one more reason you "need a lab" to do AI research. 630 lines and a single GPU is the new garage.

@SimplyGregster @GorillaOSINT They are descendants of immigrant europeans

Europeans with 0% annual GDP growth watching America become a 3rd World Country