One Flew Over the Cuckoo's Nest

España
Joined June 2020
McMurphy retweeted
F.03 just crossed 9 straight days walking on this treadmill: 348 miles and 1.2 million steps Started as a side project to stress the robot for any failures. Let's see how long it can go
289
198
60
3,846
162,887
pues quienes planearon esta guerra sabían perfectamente que acabaría conduciendo al uso de armas de destrucción masiva.
2
3
15
793
80 million foreigners don't show up by accident. This is a calculated act of war by jews, moron.
Do you know who are not looting and burning down France? The Zionists and the Jews.
156
1,344
68
5,628
47,904
McMurphy retweeted
🚨ESTO ACABA DE MATAR A ELEVENLABS 74.000 personas ya lo han guardado en GitHub y casi nadie habla de ello. Se llama VoiceStudio y es un reemplazo 100% gratis que corre en tu propio ordenador. Te permite: → Clonar tu propia voz → Doblar vídeos a 646 idiomas → Crear audiolibros y dictar → Elegir entre 17 motores Sin pago por carácter. Sin límites de uso. Todo en tu máquina. 🔖 Guárdalo.
67
398
14
2,847
196,762
A good buddy of mine did his PhD on this and now works on WiFi spatial mapping at a big tech company and It’s way more accurate than you’d imagine. You can map an entire college campus in real-time like Batman just with WiFi. Crazy stuff
breathing is enough to give you away each breath slightly changes how Wi-Fi waves reflect off your chest. with the right receiver and processing, those tiny changes can reveal your breathing rhythm the graphs in this animation use real WiRe-Breath recordings: orange is the rhythm extracted from Wi-Fi, gray is a reference breathing sensor i broke down how this works, and what else you can pull out of the radio waves around you, in the article below ↓
132
2,082
100
31,457
739,030
McMurphy retweeted
Everyone swears their phone is secretly recording their voice. You talk about buying hiking boots, and twenty minutes later Instagram serves you an ad for boots, even though your microphone permissions are turned completely off. Most of the time, ad networks don't need audio anyway. They track you through shared IP addresses, location history, and cross-site browsing graphs. At the physical hardware layer, the interesting question is whether your phone actually can listen with zero microphone permissions. Turns out you don't need a microphone when your motion sensor is literally a mechanical ear etched into silicon. 1. The microscopic machine inside your phone Inside your phone are MEMS (Micro-Electro-Mechanical Systems) accelerometers and gyroscopes. These aren't just solid-state chips. They are tiny physical machines etched directly out of silicon. An accelerometer has a microscopic silicon weight (a proof mass) suspended on silicon springs between electrical comb plates spaced barely 1.2 micrometers apart. When your phone moves, that weight shifts and changes electrical capacitance. It is sensitive enough to measure sub-nanometer deflections. 2. The acoustic impedance catch (air vs solid) Here is where the Hollywood spy movie stuff breaks down: When you talk across an open room, airborne sound waves hit your phone's glass and aluminum frame and bounce right off. Air is thin, and your phone chassis is dense metal and glass. Because of that massive acoustic impedance mismatch, over 99% of airborne sound energy simply reflects away. For a motion sensor to pick up speech, vibrations almost always have to travel through a solid medium: - The phone's own speakers: The built-in earpiece and loudspeaker bolt directly to the chassis. During a call, the speaker physically vibrates the entire circuit board from the inside. - A shared rigid surface: If your phone sits on the same wooden desk as a loud external speaker, sound conducts straight through the tabletop into the frame. 3. What researchers actually proved (Gyrophone & EarSpy) Academic papers proved this side-channel is real, but with heavy asterisks: - Stanford's Gyrophone (2014): Researchers showed smartphone gyroscopes can pick up acoustic vibrations between 80 Hz and 250 Hz (the pitch of human voices). But they didn't record someone talking across the room. They blasted audio at 75 dB from a speaker sitting on the same desk, testing a tiny 11-word vocabulary (just digits 0 to 9 and "oh"). Word recognition hit 65% for a single trained speaker, but dropped down to 26% across multiple speakers. - EarSpy (2022): Tested accelerometer readings vibrating from the phone's ear speaker during active phone calls. Because modern ear speakers are louder and sit right beside the motion sensor, they could identify the caller's gender with 98% accuracy and guess spoken digits around 56% of the time. Neither attack could transcribe an unconstrained conversation about buying shoes. 4. The quiet Android rate limit Google saw this coming and quietly patched it in Android 12 by capping motion sensor polling to 200 Hz for third-party apps, requiring a special permission (HIGH_SAMPLING_RATE_SENSORS) to poll any faster. By the Nyquist sampling theorem, a 200 Hz sampling rate cleanly captures frequencies up to 100 Hz. Adult male speech fundamentals start around 85 Hz, while female speech starts around 165 Hz, so a 200 Hz cap chops off the useful parts of speech. The catch? HIGH_SAMPLING_RATE_SENSORS is a normal install-time permission. Android grants it automatically upon install with zero pop-up warnings. Plus, 200 Hz still aliases higher frequencies back into lower readings, and researchers already showed you can fuse offset readings from the accelerometer and gyroscope to effectively double the sampling rate without the permission. 5. Why your shopping app isn't doing this today In the real world, physics makes this a nightmare for ad trackers. The moment you walk, type, tap the screen, or shift the phone in your pocket, that silicon weight moves thousands of times harder than acoustic vibrations ever could. Your physical movement creates a massive noise floor that completely drowns out speech. Ad companies don't need to burn battery processing raw vibration math when your location history, Bluetooth beacons, and credit card receipts already tell them what you want to buy. (we ain't that special mate)
41
155
22
1,091
102,724
THIS IS FUCK*NG INCREDIBLE GitHub is full of free stuff that almost nobody is using. GitHub is full of powerful free tools that most people are overlooking. Here are the 10 repositories exploding this month, ranked from the easiest to use to the most technical. What anyone can use 1. VoiceStudio: Clone voices, dub videos, and transcribe in 646 languages. Completely free and runs entirely on your own computer (57K ⭐) github.com/debpalash/VoiceSt… 2. ppt-master: Drop in a PDF or Word document and get back a real, fully editable PowerPoint with animations (59K ⭐) github.com/hugohe3/ppt-maste… 3. gods-eye-view by @bilawalsidhu: A spy satellite simulator that runs in your browser, complete with real planes, ships, and cameras on a 3D globe (50K ⭐) github.com/bilawalsidhu/gods… 4. REA: Show your AI a feature from another app and it will explain exactly how it was built, even without the source code (65K ⭐) github.com/morluto/rea 5. diagram-design by @cathrynlavery: Describe a diagram in plain words and it draws it using your own brand style, with 44 different types available (49K ⭐) github.com/cathrynlavery/dia… 6. paperclip: Run your own company of autonomous agents that work for you. This open source app lets you manage them like a real team (99K ⭐) github.com/paperclipai/paper… For making Claude Code work better 7. skills by @mattpocockuk: The small, practical skills a real engineer uses every day. Easy to adapt to your own workflow (284K ⭐) github.com/mattpocock/skills 8. ECC: A complete pack of skills, memory, and security tools that works with Claude Code, Codex, Cursor, and OpenCode (276K ⭐) github.com/affaan-m/ECC 9. andrej-karpathy-skills: One single file containing 4 rules based on what Karpathy has said models get wrong when coding (218K ⭐) github.com/multica-ai/andrej… 10. claude-mem: Saves what your agent did, summarizes it, and passes it to the next session so it never starts from zero (99K ⭐) github.com/thedotmack/claude… Share this if you found it useful ♻️ Save it before you forget 🔖
31
134
3
405
66,395
McMurphy retweeted
This is what I will be doing
1,368
1,063
308
22,258
2,200,322
🚨 رئيس Neuralink يفك لغز جدول أعمال إيلون ماسك المجهد: "ليست مجرد إدارة.. بل هندسة المبادئ الأولى!" "دي جي سيو" يوضح كيف تفكر شركات ماسك في الوقت والإنجاز من منظور فيزيائي بحت! هم يستخدمون مفهومًا يُدعى "جدول الإشارة الخضراء بالكامل": تخيل لو أن كل إشارة مرور في طريقك خضراء، لا توجد بيروقراطية، لا تأخير في الشحن، ولا عوائق إدارية.. ما هي السرعة القصوى الحقيقية لبناء شريحة أو تصميم روبوت؟ 💡 النتيجة الصادمة التي اكتشفوها: عندما تُجرد أي مشروع من "العوائق التي صنعها البشر" وتسأل عن السرعة القصوى طبقًا لقوانين الفيزياء فقط.. تكتشف أن 80% إلى 90% من الأوقات والمواعيد الضائعة هي مجرد افتراضات وتأخيرات وهمية نتوهم أنها أمور واقعية! السرعة ليست ضغطًا عشوائيًا.. بل هي نسف مستمر للتأخيرات التي تعودنا عليها وجعلناها أسلوب حياة! هل تعتقد أن شركتك أو فريق عملك يضيع 90% من الوقت في عوائق وهمية؟ 👇
🚨 سر إيلون ماسك الإداري المثير للجدل لتحقيق المستحيل! ​"أنا أضع دائماً أشد المواعيد النهائية شراسة.. احتمال تحقيقها 50% فقط، وأعلم أننا سنتأخر نصف المرات!" ​لماذا يفرضه على نفسه وشركاته؟ ​لأن هناك قانوناً في الفيزياء يشبه "تمدد الغازات" يطبق على المواعيد: لو أعطيت مشروعاً مهلة 5 سنوات (وهي زمن لا نهائي بالنسبة لماسك)، فسوف يتمدد العمل ليستهلك الـ 5 سنوات بالكامل دون أي إنجاز حقيقي! ​💡 الفلسفة الخارقة: • الموعد النهائي الشرس ليس مستحيلاً، بل يضع الفريق تحت ضغط إيجابي دائم. • بدلاً من ضياع السنوات في التراخي، يتم التركيز فوراً على العائق الفعلي وحله على وجه السرعة! ​سرعة الإنجاز ليست ضربة حظ.. بل هي فن إجبار الوقت على الانكماش! ​هل تطبق هذه الفلسفة في إدارة أهدافك ومشاريعك اليومية؟ 👇
2
32
1
182
14,123
McMurphy retweeted
It is becoming f*cking dangerous A 23-year-old dev from China just ran 300 Claude Code agents in parallel and hooked them to a verification layer so none of them could bullshit him He opens the live dashboard: - 300 agents writing and checking at the same time - Every output verified against the source in ~100 ms - Three passes over 100 company profiles → zero errors left - Everything dumped straight into an Obsidian vault The trick is the split: one model does the work, a cheaper one decides if the claim is actually true. The loop doesn’t stop until everything checks out An analyst team would bill weeks for this. He got it in one run Same kind of Second Brain + swarm setup that keeps circulating. Bookmark the full breakdown
502
806
164
10,324
3,841,778
Mistral recently released a 209M model that can predict future data locally on a CPU in ~140 milliseconds. Not every useful AI model needs to talk (just like Jev) No GPU required. It's a super small foundation model designed for time-series forecasting. Give it historical data like: 📈 sales ⚡ power usage 🏭 sensor readings 📦 demand/inventory 🌡️ telemetry …and it predicts what comes next without training a new model for each dataset. The largest version stats 🧠 209M parameters 💾 837MB FP32 🖥️ runs on CPU or GPU 📚 up to 32,768 historical time steps 🔮 forecasts up to 1,024 steps per forward pass 📊 outputs 21 probability quantiles 🎯 Mistral says it ranks #1 among pretrained/zero-shot models on all three major public forecasting benchmarks 🥇 GIFT-Eval 🥇 TIME 🥇 fev-bench It can get small like Timestral-5M → 21MB Mistral says on an 8-thread CPU: ⚡ 5M model → ~10 ms ⚡ 209M model → ~140 ms for a one-day forecast from three weeks of 15-minute data. 🔗 Links below in ALT.
43
209
25
2,508
96,467
McMurphy retweeted
The man who co-founded OpenAI turned down Meta and built a company worth $30 billion in under a year. Then he went back to his old university and told the graduates this: “The day will come when AI will do all the things that we can do.” Not some of them. All of them. University of Toronto, 6 June 2025. Ilya Sutskever gets an honorary doctorate from the school where he started. He arrived in Canada at 16, sat in a Toronto high school for about a month, and was then let into the university straight into third year. It was here, with Geoffrey Hinton and Alex Krizhevsky, that he helped build AlexNet in 2012. The neural network that kicked off the modern AI boom. Then OpenAI. Chief scientist. The research behind the models millions of people now use every day. In May 2024 he left. In June he started Safe Superintelligence. By March 2025 it was valued at $30 billion. Meta tried to buy it. He said no. So when this man stands at a podium for ten minutes, people listen. He does not talk about money. He talks about time. His argument fits in two sentences. Your brain is a biological computer. So why could a digital one not do the same things? Anything you can learn, he says, AI will be able to learn too. Slowly, or maybe not so slowly. That is the part people clip and share. The part they skip is what he tells graduates to do about it. The rules he gives are simple: Accept reality as it is. Not how you want it to be. Not how it was last year. Don’t waste time regretting the past. Every hour spent on what went wrong is an hour not spent on what comes next. Ask one question: what is the best next step? Then take it. Then ask again. You may not care about AI. AI will still affect your life. Ignoring it is not a strategy. Watch what AI can actually do. Not the hype, not the fear. The real capability. That is what tells you how much time you have. He ends by calling it the greatest challenge humanity has ever faced, and says overcoming it will bring the greatest reward. The full speech is free. Ten minutes on YouTube. No login, no course, no waitlist. What it costs you is ten minutes, and one uncomfortable question: if anything you can learn, a machine can learn too, what are you learning right now that it can’t?
17
89
9
570
168,569
China just put out a draft that could change how all of us use open-weight models. Exporting weights, checkpoints or code of the very largest models (2T+ parameters, 10^26+ FLOPs, or top 10 globally) would need government approval. API and app access stays untouched so far. If it goes through, the biggest Chinese labs may not be able to freely release their top models anymore, and teams in the US and Europe that build on them could feel it fast. My guess is this is a response to the US and its chip controls.
114
58
23
849
62,790
McMurphy retweeted
Larry Fink: I sincerely believe there is no AI bubble. “There are going to be some big failures, huge winners and some losers.” “The capital needed to build out AI is one of the great opportunities for the world.” “Countries are going to have to rebuild their grids and find new sources of energy.” “AI is going to change science and improve human lives through medicine.” “If AI can unlock abundant, cheap power, imagine how that could lift the rest of the world.”
33
17
9
153
6,099
Peter Thiel explains why he still won’t hit pause on AI, even as the labs themselves call for slowing down, and warns China’s high-tech totalitarianism can “go to the inner recesses of your thoughts” “That seems correct to me. And, you know, there’s sort of all these very complicated questions one can ask about China, but I’m, you know, I worry about totalitarianism. I don’t like it. And I think there is a certain type of, you know, high-tech totalitarianism that’s being tried out in China. It’s not as violent as, you know, Stalin or Mao or Hitler. You don’t need to kill as many people, but it can go to the inner recesses of your thoughts and it knows more about you than you know about yourself. And there’s something about that we really don’t want that to be the model that wins.” ISAACSON: “And your old friend, David Sacks, and of course, President Trump are now just very recently saying, because of the competition with China, stop talking about pausing AI, even if you think it could be an existential threat. Do you agree with that?” THIEL: “Yes. Yes. I broadly agree with that. And I’ve been friends with David Sacks for many decades. The thing that makes me less comfortable making those kinds of arguments publicly is that on some level, it’s hard when the big companies themselves are telling you to stop and to slow down.” ISAACSON: “So that’s caused you to have some pause because you were not in the pause button camp.” THIEL: “Well, I’m still not in the pause button camp. And I’m saying this on your show that I’m not in the pause button. We need to hit the accelerator as hard as possible.”
Peter Thiel says Anthropic grew out of something "almost cult-like," and asks whether a cult has ever built a company worth close to $2 trillion “And then there were, of course, parts of it that were, I don’t know, crazy and almost cult-like. But it had maybe the good dimension of a cult where there was one big thing they knew that the other people did not. And I think even though there are ways I’d like to be very critical of them today, there were ways they got that one big thing right.” “And then has a cult ever started a company as successful as Anthropic? You know, I don’t know what the most success, I mean, like even a billion dollar company seems like bigger than a crazed cult could build. So if you have a cult that’s built a company that’s worth probably close to two trillion at the IPO, it would probably be the, I don’t know, sixth, seventh largest company by market cap in the world built in like six, seven years. Has a cult ever done something like that? So, yes.” “So I want to be critical of them. And they’re things I’m uncomfortable and critical about. But the first thing I would try to do is always, you always want to steel man your opponents. You want to try to, you know, try to understand them as well as they understand themselves or even make their arguments better. And the way I would steel man the effective altruists is, first off, they got one enormously big thing right.”
2
5
2
32
9,038
McMurphy retweeted
An Anthropic researcher just told a room full of PhD students that their career plan no longer exists His 3 pieces of advice for students: 1. Use AI every day. If you're on the exponential, you'll never fall behind it. 2. Choose projects that get better when compute scales. The rest will be solved before you finish them. 3. Stop planning your career as if it were stable. The safe option no longer buys safety; it no longer exists. And a phrase that sums it all up: "Writing is thinking. If you outsource your writing, you outsource your thinking." The talk is titled "The End of the Anthropocene"
27
128
17
1,048
108,659
Peter Thiel says Anthropic grew out of something "almost cult-like," and asks whether a cult has ever built a company worth close to $2 trillion “And then there were, of course, parts of it that were, I don’t know, crazy and almost cult-like. But it had maybe the good dimension of a cult where there was one big thing they knew that the other people did not. And I think even though there are ways I’d like to be very critical of them today, there were ways they got that one big thing right.” “And then has a cult ever started a company as successful as Anthropic? You know, I don’t know what the most success, I mean, like even a billion dollar company seems like bigger than a crazed cult could build. So if you have a cult that’s built a company that’s worth probably close to two trillion at the IPO, it would probably be the, I don’t know, sixth, seventh largest company by market cap in the world built in like six, seven years. Has a cult ever done something like that? So, yes.” “So I want to be critical of them. And they’re things I’m uncomfortable and critical about. But the first thing I would try to do is always, you always want to steel man your opponents. You want to try to, you know, try to understand them as well as they understand themselves or even make their arguments better. And the way I would steel man the effective altruists is, first off, they got one enormously big thing right.”
Peter Thiel explains why aliens with faster-than-light travel must be demons or angels: only a demonic totalitarian mind meld or perfect angelic altruism could keep any one of them from launching a warp drive weapon "...If you have faster than light travel, if you have a warp drive, which is probably what you really need to cover interstellar distances, you know, what that means for military technology is that you can send weapons at warp speed and they will hit you before you see them coming. And there is no defense against a warp speed weapon." "And so it tells us that I think that if you have faster than light travel, there's something really crazy that has to be true on a cultural, political, social level. And there may be other solutions, but I'll give you my two." "You have a demonic totalitarian control of your society, where it's like you have like parapsychological mind meld with everybody, and no one can act independently of anybody else. No one can ever launch a warp drive weapon." "And then I think the other version, socially and culturally, is they have to be like perfectly altruistic, non-self-interest. They have to be angels." "It's not that they might be demons or angels, they must be demons or angels if you have faster than light travel. And both of those seem pretty crazy to me."
9
13
6
129
29,034
McMurphy retweeted
We now have a good definition of the singularity: "When AI systems speed up the pace of overall scientific and technical progress in a way that is hard for humans to comprehend." According to Jack Clark, it's already here. I would find it hard to disagree.
ANTHROPIC CO-FOUNDER TELLS THE IMF THE SINGULARITY HAS ALREADY BEGUN "I believe that this year, we've gained convincing evidence that it has now begun" "a point where our models must be discarded and a new reality rules" "the singularity is here" at least this time Clark shares his definition of what "singularity" means, and even by this definition I'm not sure we are there yet. AI is clearly speeding up progress dramatically in some areas, but it is still uneven, with plenty of bottlenecks preventing broad, self-sustaining acceleration even the recent math progress still looks more like a massive speed-up of technical work than completely new theories or brilliant conceptual breakthroughs. so if you ask me, the singularity is not here yet, but the current pace may already be enough to create massive social and technological turbulence
42
58
5
645
25,399
Dicen que #Málaga tiene un sol especial. Durante varios días y meses estaré caminando la ciudad para capturar solo la ciudad y su luz. Construiré este hilo sin prisa, solo con imágenes. Es una pequeña carta de amor para Málaga. Espero les guste. 🔽
9
72
9
399
72,595