Injection for Xcode @Injection4Xcode

Seriously recommend following this guy. Great post after great post.

Ihtesham Ali @ihtesham2005

5 days ago

A neuroscientist at Johns Hopkins spent 30 years proving that the human brain was designed to do its best work hungry, and the modern habit of eating every few hours is the reason most people feel slow and foggy at the desk. His name is Mark Mattson. He spent two decades

33 136 582 41K 604

📣 Vous êtes nombreux à me demander ce qu'il se passe à la @CourPenaleInt, avec son Procureur, Karim Khan. Entre accusations, soupçons, conclusions contradictoires et une possible révocation, difficile de s'y retrouver. 🧵Petit thread chronologique. ⬇️ lemonde.fr/international/…

People always say Iran became the world's problem after the Islamic revolution in 1979. Where things went off the rails is further back after the British and American lead coup d'état of 1953. Churchill's last legacy of the colonial way of doing things. en.wikipedia.org/wiki/1953_Iran…

If you're using #InjectionIII or #InjectionNext on iOS 27 there are new release candidates available that fix an alignment problem in fishhook. github.com/johnno1962/fis…

open models. ❤️

US sanctions. Here it what you need to know. In the clearest words I have on the matter.

Courage International AISBL @couragefound

a week ago

Francesca Albanese has been sanctioned for doing her job and demanding an end to the Genocide in Palestine. The EU has failed to protect her, the UN has failed to protect her. In this video message .@FranceskAlbs updates us on the current state of affairs regarding the US

64 1K 2K 214K 159

@vatsal_manot Hi! Loved @diagnose to force warnings or suppress warnings at the site x.com/bErKut89/statu…

Vlad Berkuta @bErKut89

2 months ago

@Injection4Xcode “Large Language Brother”

0 0 1 164 0

Has anybody coined the term "Large Brother" for the upcoming widespread use of LLMs in surveillance?

If you're having difficulty setting up macOS 27 in a VM (eg UTM)

Lots of good stuff in there developer.apple.com/videos/play/ww…

A TV writer with no philosophy degree read Aristotle, Kant, Bentham, and Scanlon back to back, built a sitcom around what he found, then explained in one book why the trolley problem is no longer a thought experiment and the people who need to understand this most are the ones building AI. His name is Michael Schur. He co-created Parks and Recreation and Brooklyn Nine-Nine. But The Good Place was the project that broke him open. On the surface it was a comedy about a woman who accidentally ends up in heaven. Underneath, it was a philosophy seminar. Every episode was built around a real ethical framework. He had to actually understand all of it to make any of it funny. After the show ended, he wrote the book anyway. He called it "How to Be Perfect." It begins with the most honest opening line in any philosophy book ever written: Should I punch my friend in the face for no reason? No. That is not a joke. That is his method. He starts with the obvious and builds toward the impossible. Here is the framework he built, and why the most dangerous people in tech right now are running exactly one of the four schools of thought without knowing any of the others exist. The first school is Virtue Ethics. Aristotle built it around 350 BCE. The question it asks is not "what should I do?" It asks "what kind of person should I be?" The idea: become genuinely good, and good actions will follow naturally. You build courage. You build honesty. You build practical wisdom. Then you trust the person you built. The second school is Deontology. Kant built it in the 18th century and it is the exact opposite. Kant did not care about the person. He cared about the rule. His version: act only in ways you would be comfortable turning into a universal law. If everyone lied whenever it was convenient, the concept of truth would dissolve. So you never lie. Even if the truth gets someone killed. The rule is absolute because the moment you make one exception, it stops being a rule. The third school is Utilitarianism. This is the one that should stop anyone building AI cold. Jeremy Bentham invented it in the late 1700s. The principle sounds beautiful: the right action is whichever one produces the greatest good for the greatest number of people. Pure math. Pure outcome. Intention means nothing. Only consequences count. Schur runs it through the trolley problem, the most famous thought experiment in philosophy. You are driving a runaway trolley. Five people are tied to the main track. You can pull a lever and redirect it to a side track where only one person is tied. Do nothing and five die. Pull the lever and you kill one to save five. A utilitarian says pull the lever. The math is obvious. Now the same problem with one change. You are on a bridge above the track. A large man is standing next to you. The physics are clear: if you push him off the bridge, his body stops the trolley. Five people live. He dies. The math is identical. Almost nobody will push the man. Even people who pulled the lever instantly in the first version refuse. The utilitarian has no answer for why these two situations feel different. The numbers are the same. The outcome is the same. The only thing that changed is whether you are using another human being as a tool. That gap between the math being correct and the action feeling monstrous is exactly where AI ethics collapses every single time. The fourth school is Contractualism, built by the philosopher T.M. Scanlon. It asks the question that Kant's rules and Bentham's math both miss. What principles could be justified to everyone affected? Not the majority. Not the person with the most power. Everyone. Including the one person who ends up on the shorter end of the calculation. Schur's conclusion is the part that people who live inside growth frameworks and optimization loops will resist the hardest. None of the four schools is correct on its own. Each one has a scenario where following it perfectly produces something most humans recognize as evil. Pure utilitarianism justifies harvesting one person's organs to save five dying patients. Pure deontology says you cannot lie to the murderer asking where your friend is hiding. Each system, taken to its logical extreme, becomes a machine that produces monsters while generating perfect internal justification for doing so. The way out is not picking the right framework and following it harder. The way out is using all four as lenses. Ask what Aristotle would do. Ask what Kant would allow. Do the utilitarian math. Then ask Scanlon's question: could you justify this to the person it hurts most? Where those four answers overlap, you are probably on solid ground. Where they pull in different directions, you are in territory that deserves far more than a two-hour board meeting. Schur also coined a term that has been stuck in my head since I finished the book. Moral Exhaustion. The feeling of living in an age where you can know, in real time, every ethical implication of every product you use, every company you work for, every piece of code you ship. The gap between what you know and what you can actually change becomes so large that the easiest response is to stop asking. He says that response is understandable. He also says that choosing not to ask is itself a moral choice, and the consequences of that choice scale in exact proportion to the power you hold. A person building a product one billion people will use is not operating at the scale where shrugging is a neutral act. The people who built the most consequential technologies of the last decade were not evil. Most were genuinely trying to do good. They ran the utilitarian math. They saw a billion users. They saw engagement numbers that looked like impact. They optimized for the greatest good for the greatest number and did not notice until much later that the people being turned into variables in the math were still people. Schur read 2,500 years of philosophy and the lesson he came out with fits in one sentence. You cannot use a single framework because every single framework, followed perfectly, eventually produces the wrong answer. The people who cause the most damage are not the ones who do not care about ethics. They are the ones who found one framework they liked, felt good about it, and stopped asking. The trolley problem is not a thought experiment anymore. It runs on servers. It gets optimized overnight. And the people making those decisions right now have never once asked what Scanlon would say.

@OpenAIDevs Has nobody seen #InjectionIII/#InjectionNext? (works on a device, MCPs your actual screen and can send taps) - Hire me for a while! github.com/johnno1962/Inj…

Just plugged in a 20 year iPod nano (MA489LL). No lag. No loading indicators. UI is practically instant. No touchscreen. I plugged in my Bose headphones and played Black Eyed Peas like 2006 was yesterday. Just fucking worked. Incredible. Technology has gotten worse in so many ways. Younger people have no idea.

@neeratanden dem incompetence led us to trump twice. the left has never stopped opposing trump. this insane convo only exists on this fucking website. most ppl are mad at dems for not doing enough. all of the spoxppl for the party on this stupid website are out of touch w the masses.

A McKinsey consultant with no PhD, no AI background, and no academic position quietly built the most-watched deep learning course on Earth and gave the entire thing away for free. I opened the first lesson at 1am and could not believe a single human had taught this many people without ever charging a cent for it. His name is Jeremy Howard. The course is called "Practical Deep Learning for Coders" For most of his career, Jeremy Howard had no business teaching artificial intelligence to anyone. He spent 8 years as a management consultant at McKinsey and AT Kearney. He started an email company called FastMail in Australia and ran it for years. He built an insurance pricing startup that got acquired by Lexis-Nexis. None of that was AI work. None of it was academic. He did not have a doctorate in computer science or mathematics or anything else. He was a businessman who happened to be very good at writing code. Then in 2010 and 2011, he became the top ranked competitor in data science contests on Kaggle, beating teams of PhDs from the most credentialed labs in the world. He was eventually made president of the company. And what he saw from that seat was the thing that ended up changing his life. The best deep learning research on Earth was being done by maybe a few hundred people in five or six elite labs in San Francisco, London, and Toronto. To get into those labs, you needed a PhD from Stanford or MIT, a recommendation from a tenured professor, and access to expensive GPUs that no individual could afford. The practical knowledge of how to actually train these models was almost never written down. It lived inside the heads of a small priesthood, and the priesthood was almost entirely closed to outsiders. In 2016 Jeremy Howard and Rachel Thomas decided to break that gate. They founded fast.ai with one mission. Take the techniques that the elite labs were using and teach them to anyone in the world who could write a Python loop. No PhD required. No advanced math required. No expensive hardware required. Just a laptop, an internet connection, and the willingness to actually finish the lessons. The way they did it was the part that almost nobody in academia had ever tried. Every other deep learning course on the planet started with theory. You learned linear algebra. You learned multivariate calculus. You learned probability theory. You spent 6 months on the foundations before you ever got near a working neural network. By the time most students reached the part where they could actually build something, they had quit. Howard inverted the entire curriculum. Lesson one of his course is not theory. It is a working image classifier that you train in 15 minutes on a model that can distinguish dog breeds with 99 percent accuracy. You build the thing first. You make it work first. Only then, once you have proven to yourself that you can actually do this, do you start peeling back the layers to understand why it works. His justification was simple. The reason most people quit learning hard things is not that the material is too difficult. It is that the curriculum is structured to make them feel stupid for as long as possible before they ever get to do anything interesting. Howard refused to do that to his students. He believed that if you could see something working with your own hands on day one, you would have the motivation to fight through the hard math three months later. And he was right. The course has now been viewed over 6 million times. His students include a Canadian dairy farmer who used the course to build an AI system to monitor the health of his goats. They include a French math teacher and a network of doctors in Africa. They include people who walked into the lessons knowing nothing about AI and walked out building production systems at Google Brain, OpenAI, Adobe, Amazon, and Tesla. In 2018, Jeremy Howard and a young researcher named Sebastian Ruder published a paper called "Universal Language Model Fine-tuning." It introduced a transfer learning technique for language models that worked so well it cut error rates on text classification by 24 percent on the hardest benchmarks in the field. That technique, refined and scaled by the labs that came after, became the foundation of how every major language model on Earth is trained today. ChatGPT. Claude. Gemini. The fine-tuning step that makes them useful traces back to the methodology in that paper. The man who co-wrote it had no PhD. He had been teaching the same ideas to strangers on the internet for free a year before he ever published the paper. His students were already building with it before the elite labs had even read it. The best things on the internet are almost never the ones with paywalls in front of them. They are the ones built by people who decided the gate was the problem and then quietly walked around it.

@unusual_whales "once, men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them." Dune, 14.

Thank you @AOC for stating the obvious: the anti-tax obsession is effectively a tax on young American workers just starting out in the economy.

NOTUS @NOTUSreports

2 months ago

The differences between Rep. AOC’s and Sen. Bernie Sanders’ approach to tax cut plans are indicative of the divides Democrats must bridge as they form their agenda going into 2028. notus.org/economy/democr…

7 26 224 511K 88

Nobody could figure out why the abandoned Hendricks apple orchard suddenly bloomed in April 2019. The trees hadn't produced fruit in eleven years. County agriculture office sent two inspectors. They found sixty thousand honeybees working the property - a massive colony that had escaped from Tomás Vega's apiary three miles south. Tomás had reported the swarm missing in March. He expected them dead. Instead they'd colonized the hollow barn on the Hendricks lot and cross-pollinated every surviving tree. That October, the orchard produced twenty-two tons of Cortland apples. The Hendricks family offered Tomás a permanent lease. He moved his entire operation there the following spring.

Trump, the biggest failed strategist in US history, has plunged America into a strategic catastrophe. During his self-proclaimed “victory parade” against Iran, he squandered at least 45% of the US's precision-guided missile arsenal in just seven weeks—including half of all THAAD missiles and nearly 50% of Patriot interceptor missiles. This isn't some fake news blog reporting this, but CNN, citing a CSIS analysis and internal Pentagon data. The result? An “imminent risk” of munitions depletion should a real conflict erupt in the coming years—for example, with China. Trump has ruined the US defense capability for years to come. And for what? For nothing. No regime change in Iran. No destroyed nuclear program. No strategic breakthrough. Just a shaky ceasefire that gives the mullahs time to rearm while America stands naked. Trump, the great “Art of the Deal” master, has once again only produced hot air – and in doing so, burned through the most expensive and scarce weapons in the USA like a pubescent boy with fireworks.