Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

the Deepfake Offensive Toolkit dot (aka Deepfake Offensive Toolkit) makes real-time, controllable deepfakes ready for virtual cameras injection. dot is created for performing penetration testing against e.g. identity verification. #infosec #redteam

HackGit

53,873 subscribers

75,672 Aufrufe • vor 2 Jahren •via X (Twitter)

Wissenschaft & Technologie #infosec #redteam

Anya Rossi• Live Now

Private livecam show

5 Kommentare

Profilbild von iM4n Emperor

iM4n Emperorvor 2 Jahren

بره تمایلاین

Profilbild von LessA

LessAvor 2 Jahren

@AzeTIIx

Profilbild von Constantin Claudiu

Constantin Claudiuvor 2 Jahren

@SaveToNotion #thread #aiiii

Profilbild von CyberPunk

CyberPunkvor 2 Jahren

😂😂

Profilbild von ProjHastings

ProjHastingsvor 2 Jahren

This is a delight

Ähnliche Videos

someone built a tool that puts ANY FACE on your webcam in REAL TIME called DOT - deepfake offensive toolkit, open source, works right now dot injects a real-time deepfake into your virtual camera zoom, google meet, teams, any video call app sees the fake face as your real webcam it ships with multiple deepfake engines (FOMM, SimSwap, SimSwapHQ), you pick the model based on your GPU and how real you need it to look built by sensity ai for red teams to test if your identity verification actually holds up against deepfakes

someone built a tool that puts ANY FACE on your webcam in REAL TIME called DOT - deepfake offensive toolkit, open source, works right now dot injects a real-time deepfake into your virtual camera zoom, google meet, teams, any video call app sees the fake face as your real webcam it ships with multiple deepfake engines (FOMM, SimSwap, SimSwapHQ), you pick the model based on your GPU and how real you need it to look built by sensity ai for red teams to test if your identity verification actually holds up against deepfakes

chiefofautism

37,088 Aufrufe • vor 4 Monaten

#Deepfakes are videos, images, even sounds artificially generated by #machinelearning For the first time, a virtually aged deepfake of a 9-year-old child has been created so that she can act and argue like an adult woman. Ella is representative of an entire generation of children. Experts predict that by 2030, two-thirds of all cases of identity theft will occur in this way. Possible consequences include profiling by data brokers, hacking, facial recognition, pedophilia and others Credits: Deutsche Telekom

#Deepfakes are videos, images, even sounds artificially generated by #machinelearning For the first time, a virtually aged deepfake of a 9-year-old child has been created so that she can act and argue like an adult woman. Ella is representative of an entire generation of children. Experts predict that by 2030, two-thirds of all cases of identity theft will occur in this way. Possible consequences include profiling by data brokers, hacking, facial recognition, pedophilia and others Credits: Deutsche Telekom

Tariq Malik ™

58,521 Aufrufe • vor 3 Jahren

Meanwhile the Ukrainians revealed that the counter-offensive is nothing more than a “psychological operation against the aggressors” and all this time they were just trolling everyone in order to buy themselves time for the real counteroffensive. Can you imagine the grief of the family of a dead soldiers near Zaporozhye or Avdeevka, who learned that not only did his command destroy him just like that, for the sake of intimidating Muscovites and for fun, but they would not receive any payments for this, since there was a counter-offensive and it wasn’t there at all?

Meanwhile the Ukrainians revealed that the counter-offensive is nothing more than a “psychological operation against the aggressors” and all this time they were just trolling everyone in order to buy themselves time for the real counteroffensive. Can you imagine the grief of the family of a dead soldiers near Zaporozhye or Avdeevka, who learned that not only did his command destroy him just like that, for the sake of intimidating Muscovites and for fun, but they would not receive any payments for this, since there was a counter-offensive and it wasn’t there at all?

Olga Bazova

245,216 Aufrufe • vor 2 Jahren

With the Kentucky Derby coming this weekend, it's a perfect time for all you horse racing enthusiasts to check out Photo Finish™ LIVE 🐎 Virtual Horse Racing if you haven't yet! Photo Finish™ LIVE 🐎 Virtual Horse Racing is partnered with Churchill Downs, and the official Virtual Kentucky Derby also runs this Saturday! In Photofinish game you own & breed virtual horses that compete against others for real $$$ (Virtual Kentucky Derby purse is $10,000). It's an incredible game that makes the horse racing hobby more accessible to all.

With the Kentucky Derby coming this weekend, it's a perfect time for all you horse racing enthusiasts to check out Photo Finish™ LIVE 🐎 Virtual Horse Racing if you haven't yet! Photo Finish™ LIVE 🐎 Virtual Horse Racing is partnered with Churchill Downs, and the official Virtual Kentucky Derby also runs this Saturday! In Photofinish game you own & breed virtual horses that compete against others for real $$$ (Virtual Kentucky Derby purse is $10,000). It's an incredible game that makes the horse racing hobby more accessible to all.

Isenhart Stables 🐎🌹 $CROWN

29,040 Aufrufe • vor 1 Jahr

🚨 ALERT: I've implemented a system for near-instant deepfakes (🎥 Video & 🎤 Audio) — we’re talking LIVE Webcam/Microphone to near-real-time Deepfake Video/Voice. Tomorrow, someone else will. Are we ready for a world where seeing isn’t believing? 🤔 This isn’t sci-fi anymore. The tech to manipulate audio and video in near-real-time exists now. I’ve built upon it, used it in real-time meetings, and I’m sure others have or will soon too. We need urgent action: 🧠 Teach digital skepticism 🛠️ Develop better detection tools (I’m actively working on this) This isn’t future tech—I’ve used it in actual meetings. It’s not about fear—it’s about preparedness. What is your organization doing to address this?

🚨 ALERT: I've implemented a system for near-instant deepfakes (🎥 Video & 🎤 Audio) — we’re talking LIVE Webcam/Microphone to near-real-time Deepfake Video/Voice. Tomorrow, someone else will. Are we ready for a world where seeing isn’t believing? 🤔 This isn’t sci-fi anymore. The tech to manipulate audio and video in near-real-time exists now. I’ve built upon it, used it in real-time meetings, and I’m sure others have or will soon too. We need urgent action: 🧠 Teach digital skepticism 🛠️ Develop better detection tools (I’m actively working on this) This isn’t future tech—I’ve used it in actual meetings. It’s not about fear—it’s about preparedness. What is your organization doing to address this?

@Tech

13,403 Aufrufe • vor 1 Jahr

This is so cool. Tesla can take footage from its massive vehicle fleet and synthetically create new driving scenarios to test edge cases and improve safety for its self-driving software. Tesla can also stitch footage from all 8 cameras into a fully drivable 3D environment—letting engineers steer, brake, and navigate as if they were on real roads, all powered by neural network–generated video streams. • Can simulate 8 Tesla camera feeds simultaneously — fully synthetic. • Used for testing, training, and reinforcement learning. • Allows adversarial event injection (e.g., adding a pedestrian or vehicle cutting in). • Enables replaying past failures to verify new model improvements. • Can run in near real-time, letting testers “drive” inside a simulated world.

This is so cool. Tesla can take footage from its massive vehicle fleet and synthetically create new driving scenarios to test edge cases and improve safety for its self-driving software. Tesla can also stitch footage from all 8 cameras into a fully drivable 3D environment—letting engineers steer, brake, and navigate as if they were on real roads, all powered by neural network–generated video streams. • Can simulate 8 Tesla camera feeds simultaneously — fully synthetic. • Used for testing, training, and reinforcement learning. • Allows adversarial event injection (e.g., adding a pedestrian or vehicle cutting in). • Enables replaying past failures to verify new model improvements. • Can run in near real-time, letting testers “drive” inside a simulated world.

Sawyer Merritt

591,318 Aufrufe • vor 8 Monaten

The "three fingers in front of your face" test (also called the 3-Finger Test) is a simple, viral trick popularized in early 2026 for spotting deepfakes or AI-generated video during live calls, especially with scammers. How It Works On a video call (Zoom, FaceTime, etc.), you ask the other person: "Can you hold up three fingers in front of your face?" A real human does this easily — their hand overlaps their face naturally. Many (especially older) deepfake or real-time face-swap systems struggle with this because: - They have trouble rendering occlusion (when the hand blocks part of the face) accurately in real time. - This often causes visible glitches: blurring, warping, distortion, unnatural blending, or the AI refusing / hesitating because it can't process the complex interaction properly. In the original viral clip (from cybersecurity researchers like Jim Browning and Huntress Labs), a suspected scammer hesitates, deflects ("That's too much..."), partially complies poorly, or the video glitches — exposing the fake. Why It Fits the Cognitive Wild West This test is a practical example of the "Cognitive Wild West" idea: in an era of synthetic media, people are developing quick hacks and personal "defenses" because there's no widespread, reliable system for verifying reality. It's low-tech cognitive security against high-tech deception — part of the arms race between AI creators and everyday users trying not to get scammed or manipulated. Limitations (It's Not Foolproof) Already becoming obsolete — Newer, more advanced deepfake models (as of mid-2026) handle hand-face interactions much better. Scammers now know the trick and can train around it or use better tech. It works best on real-time generated deepfakes, not pre-recorded or highly polished ones. False confidence is a risk: relying on one trick can make you less safe if you ignore other red flags (odd lighting, unnatural eye movement, audio mismatches, etc.). Better Practices in the Wild West Security experts recommend combining it with: - Asking for unpredictable actions (e.g., specific movements or objects). - Verifying identity through other channels (call back on a known number, check details independently). - Using emerging detection tools or just defaulting to higher scepticism on unsolicited video calls. It's a fun, memorable meme in the broader discussion of living with AI-generated content, but like everything in this space, it's temporary — the frontier keeps moving.

The "three fingers in front of your face" test (also called the 3-Finger Test) is a simple, viral trick popularized in early 2026 for spotting deepfakes or AI-generated video during live calls, especially with scammers. How It Works On a video call (Zoom, FaceTime, etc.), you ask the other person: "Can you hold up three fingers in front of your face?" A real human does this easily — their hand overlaps their face naturally. Many (especially older) deepfake or real-time face-swap systems struggle with this because: - They have trouble rendering occlusion (when the hand blocks part of the face) accurately in real time. - This often causes visible glitches: blurring, warping, distortion, unnatural blending, or the AI refusing / hesitating because it can't process the complex interaction properly. In the original viral clip (from cybersecurity researchers like Jim Browning and Huntress Labs), a suspected scammer hesitates, deflects ("That's too much..."), partially complies poorly, or the video glitches — exposing the fake. Why It Fits the Cognitive Wild West This test is a practical example of the "Cognitive Wild West" idea: in an era of synthetic media, people are developing quick hacks and personal "defenses" because there's no widespread, reliable system for verifying reality. It's low-tech cognitive security against high-tech deception — part of the arms race between AI creators and everyday users trying not to get scammed or manipulated. Limitations (It's Not Foolproof) Already becoming obsolete — Newer, more advanced deepfake models (as of mid-2026) handle hand-face interactions much better. Scammers now know the trick and can train around it or use better tech. It works best on real-time generated deepfakes, not pre-recorded or highly polished ones. False confidence is a risk: relying on one trick can make you less safe if you ignore other red flags (odd lighting, unnatural eye movement, audio mismatches, etc.). Better Practices in the Wild West Security experts recommend combining it with: - Asking for unpredictable actions (e.g., specific movements or objects). - Verifying identity through other channels (call back on a known number, check details independently). - Using emerging detection tools or just defaulting to higher scepticism on unsolicited video calls. It's a fun, memorable meme in the broader discussion of living with AI-generated content, but like everything in this space, it's temporary — the frontier keeps moving.

9/11 Revisited

2,330,960 Aufrufe • vor 17 Tagen

🚨THIS SHOULD WORRY EVERY SINGLE AMERICAN (AND NON-AMERICAN) IN THIS COUNTRY AND QUITE FRANKLY, THE ENTIRE GLOBE. THEY WILL STOP AT NOTHING TO BUILD A SURVEILLANCE STATE. IT HAS ALREADY BEGUN.🚨 Google paid $32 billion for Wiz. Not for the tech. For the team. 👀 Wiz founders: Assaf Rappaport, Yinon Costica, Roy Reznik, Ami Luttwak. All Unit 8200 veterans. Unit 8200 is the NSA of Israel. Elite military cyber intelligence. And they are now running Google Cloud security architecture. For every enterprise. Every government contract. Every hospital. Every bank on that platform. 😤 Let's talk about what Unit 8200 actually built. Pegasus spyware. NSO Group. Documented. Used to surveil journalists, activists, and heads of state. Jamal Khashoggi. The Washington Post journalist murdered inside a Saudi consulate. Unit 8200 surveillance infrastructure was embedded in the kill chain. That is not a theory. That is what the reporting shows. 💀 Now same alumni network. Different company. Authentics. That is the identity verification layer behind your X account verification. Your passport photo. Your facial biometric scan. Your government-issued ID. Sitting in a database. Founded by Unit 8200 alumni. Look it up. www. au10tix dot com. That is the actual company. Rebranded. Still the same origin. 😒 This is not one company‼️ This is an intelligence alumni network building the identity and security infrastructure of the American internet. 🌐 😤 Wiz = cloud security Au10tix/Authentics = identity verification NSO Group = offensive cyber Cellebrite = device extraction (law enforcement tool used in the US) All Unit 8200. All Israeli military cyber intelligence graduates. All now embedded inside American critical infrastructure. Ask yourself one question. If Chinese military cyber unit veterans founded the company holding your biometric passport scan and running Google Cloud security... ...would that be front page news? You already know the answer. Receipts don't have a religion or feelings. Do you think this is good for Americans' and our constitution? Sound off below 👂

🚨THIS SHOULD WORRY EVERY SINGLE AMERICAN (AND NON-AMERICAN) IN THIS COUNTRY AND QUITE FRANKLY, THE ENTIRE GLOBE. THEY WILL STOP AT NOTHING TO BUILD A SURVEILLANCE STATE. IT HAS ALREADY BEGUN.🚨 Google paid $32 billion for Wiz. Not for the tech. For the team. 👀 Wiz founders: Assaf Rappaport, Yinon Costica, Roy Reznik, Ami Luttwak. All Unit 8200 veterans. Unit 8200 is the NSA of Israel. Elite military cyber intelligence. And they are now running Google Cloud security architecture. For every enterprise. Every government contract. Every hospital. Every bank on that platform. 😤 Let's talk about what Unit 8200 actually built. Pegasus spyware. NSO Group. Documented. Used to surveil journalists, activists, and heads of state. Jamal Khashoggi. The Washington Post journalist murdered inside a Saudi consulate. Unit 8200 surveillance infrastructure was embedded in the kill chain. That is not a theory. That is what the reporting shows. 💀 Now same alumni network. Different company. Authentics. That is the identity verification layer behind your X account verification. Your passport photo. Your facial biometric scan. Your government-issued ID. Sitting in a database. Founded by Unit 8200 alumni. Look it up. www. au10tix dot com. That is the actual company. Rebranded. Still the same origin. 😒 This is not one company‼️ This is an intelligence alumni network building the identity and security infrastructure of the American internet. 🌐 😤 Wiz = cloud security Au10tix/Authentics = identity verification NSO Group = offensive cyber Cellebrite = device extraction (law enforcement tool used in the US) All Unit 8200. All Israeli military cyber intelligence graduates. All now embedded inside American critical infrastructure. Ask yourself one question. If Chinese military cyber unit veterans founded the company holding your biometric passport scan and running Google Cloud security... ...would that be front page news? You already know the answer. Receipts don't have a religion or feelings. Do you think this is good for Americans' and our constitution? Sound off below 👂

Danks

76,942 Aufrufe • vor 4 Monaten

Worldcoin is a game-changer in crypto and Ai world!!! Let me get straight and tell you more about the World and their $WLD ecosystem They’re not just another project—they’re paving the way for mass adoption with some insane innovations Here's how 👇 World App: This isn’t your typical wallet. It’s crypto’s first super app! A non-custodial Ethereum wallet with Mini Apps, where you can manage assets, verify with World ID, and access third-party services—all from one place, securely. World ID: In an AI-dominated world, privacy matters. World ID is already the largest identity protocol, with millions of verified users across 160+ countries. Their biometric verification keeps your data private and protected. Worldchain: Just launched and already massive. Designed to scale with a global user base, it’s the largest layer-2 chain on day one, enabling secure, human-prioritized transactions and on-chain identity verification. Orb 2.0: Powered by NVIDIA, this next-gen biometric scanner is 5x faster and better at fraud detection. Real-time, privacy-preserving verification at its best! $WLD is redefining what’s possible in Web3 by solving real problems like deepfakes and digital fraud. Sam Altman and the team are pushing boundaries and making this space future-proof

Worldcoin is a game-changer in crypto and Ai world!!! Let me get straight and tell you more about the World and their $WLD ecosystem They’re not just another project—they’re paving the way for mass adoption with some insane innovations Here's how 👇 World App: This isn’t your typical wallet. It’s crypto’s first super app! A non-custodial Ethereum wallet with Mini Apps, where you can manage assets, verify with World ID, and access third-party services—all from one place, securely. World ID: In an AI-dominated world, privacy matters. World ID is already the largest identity protocol, with millions of verified users across 160+ countries. Their biometric verification keeps your data private and protected. Worldchain: Just launched and already massive. Designed to scale with a global user base, it’s the largest layer-2 chain on day one, enabling secure, human-prioritized transactions and on-chain identity verification. Orb 2.0: Powered by NVIDIA, this next-gen biometric scanner is 5x faster and better at fraud detection. Real-time, privacy-preserving verification at its best! $WLD is redefining what’s possible in Web3 by solving real problems like deepfakes and digital fraud. Sam Altman and the team are pushing boundaries and making this space future-proof

Lau

116,744 Aufrufe • vor 1 Jahr

Trevor Wittman Says Justin Gaethje Will Be More Offensive With His Grappling Against Paddy Pimblett 🥊👀 "We're ready for everything. We did not go into this fight thinking it's going to be just a striking fight at all. The proudest thing with Justin is the amount of time he's put into his grappling because his wrestling is so high level. When we had the Oliveira fight, we were defensive, and you can't win fights like that. You have to win offensively, and Justin is one of those guys, when you grapple with him, his ground and pound is so unique." (Via ESPN MMA )

Trevor Wittman Says Justin Gaethje Will Be More Offensive With His Grappling Against Paddy Pimblett 🥊👀 "We're ready for everything. We did not go into this fight thinking it's going to be just a striking fight at all. The proudest thing with Justin is the amount of time he's put into his grappling because his wrestling is so high level. When we had the Oliveira fight, we were defensive, and you can't win fights like that. You have to win offensively, and Justin is one of those guys, when you grapple with him, his ground and pound is so unique." (Via ESPN MMA )

Red Corner MMA

14,024 Aufrufe • vor 6 Monaten

🌌 YogaPetz Is Finally Here ✨ The wait is over. Your NFT is no longer static — it lives, breathes, and performs. 🌀 The World of YogaPetz From peaceful towns to mystical landscapes, 4 tribes emerge: 🧍 Human | 🐾 Petz | 👁 Chakras | 🧚🏻‍♀️‍ Elf Each one carries unique traits and stories — all brought to life in high-fidelity 3D. 🎨 What’s Inside Your NFT? ✔ Fully rigged 3D model (body + face) ✔ well defined blendshapes for emotional expression ✔ Physics-enabled for realistic hair, clothing, and accessories ✔ Optimized for real-time use and production pipelines 🎯 Use Cases: 🎥 Livestream-ready VTuber or persona 🎮 NPCs for games or interactive apps 🎞 Create animated shorts or TikToks 🎤 Host virtual concerts or digital events 🏷 Launch your brand’s virtual ambassador 🤖 Coming soon: Integrate with WELLAIOS AI Agents for autonomous digital identity 🚀 Reveal your YogaPetz Egg and explore the new era of 3D NFTs: 👉 #YogaPetz #WELLAIOS

WELL3

16,442 Aufrufe • vor 1 Jahr

XBOW raised $117M to build AI hacking agents. Now someone just open-sourced it for FREE. Strix deploys autonomous AI agents that act like real hackers - they run your code dynamically, find vulnerabilities, and validate them through actual proof-of-concepts. Why it matters: The biggest problem with traditional security testing is that it doesn't keep up with development speed. Strix solves this by integrating directly into your workflow: ↳ Run it in CI/CD to catch vulnerabilities before production ↳ Get real proof-of-concepts, not false positives from static analysis ↳ Test everything: injection attacks, access control, business logic flaws The best part? You don't need to be a security expert. Strix includes a complete hacker toolkit - HTTP proxy, browser automation, and Python runtime for exploit development. It's like having a security team that works at the speed of your CI/CD pipeline. The best part is that the tool runs locally in Docker containers, so your code never leaves your environment. Getting started is simple: - pipx install strix-agent - Point it at your codebase (app, repo, or directory) Everything is 100% open-source! I've shared link to the GitHub repo in the replies!

XBOW raised $117M to build AI hacking agents. Now someone just open-sourced it for FREE. Strix deploys autonomous AI agents that act like real hackers - they run your code dynamically, find vulnerabilities, and validate them through actual proof-of-concepts. Why it matters: The biggest problem with traditional security testing is that it doesn't keep up with development speed. Strix solves this by integrating directly into your workflow: ↳ Run it in CI/CD to catch vulnerabilities before production ↳ Get real proof-of-concepts, not false positives from static analysis ↳ Test everything: injection attacks, access control, business logic flaws The best part? You don't need to be a security expert. Strix includes a complete hacker toolkit - HTTP proxy, browser automation, and Python runtime for exploit development. It's like having a security team that works at the speed of your CI/CD pipeline. The best part is that the tool runs locally in Docker containers, so your code never leaves your environment. Getting started is simple: - pipx install strix-agent - Point it at your codebase (app, repo, or directory) Everything is 100% open-source! I've shared link to the GitHub repo in the replies!

Akshay 🚀

96,623 Aufrufe • vor 8 Monaten

To everyone wondering if Tesla's FSD moat has eroded thanks to Nvidia's keynote - here's the answer: Think of Nvidia as a seller of a toolkit. You can buy pieces built specifically for a job but once you have the tools you still have to build the entire project. In this case...a model Many seem to be in awe over Nvidia's Cosmos which is a platform that includes World Foundation Models that can generate synthetic data for training AI systems like autonomous vehicles I'll go into much more depth in today's episode but here's a clip of Ashok Elluswamy at CVPR '23 explaining how Tesla is already using a similar approach to augment its real world data set This also doesn't even consider the fact that much of the auto industry using Nvidia "tools" will be forced to pay 50%+ margins just to buy the toolkit and will be locked in to Nvidia's system Nor does this touch on legacy auto needing to hire top ML engineering talent to actually put these tools to work The path to autonomy will be real world data as the foundation and simulations/synthetic data as a supplement. There is no path to autonomy with synthetic data alone. More to come later $TSLA As Elon said earlier this year, "it's remarkable how quickly we run out of human-created data. Reality itself and synthetic data ftw" "What you are seeing here is purely generated video sequences - given the past videos the network predicts some sample from the future, hopefully the most likely sample. It is being predicted not just for one camera, but it predicts for all 8 cameras around the car jointly" - Ashok

To everyone wondering if Tesla's FSD moat has eroded thanks to Nvidia's keynote - here's the answer: Think of Nvidia as a seller of a toolkit. You can buy pieces built specifically for a job but once you have the tools you still have to build the entire project. In this case...a model Many seem to be in awe over Nvidia's Cosmos which is a platform that includes World Foundation Models that can generate synthetic data for training AI systems like autonomous vehicles I'll go into much more depth in today's episode but here's a clip of Ashok Elluswamy at CVPR '23 explaining how Tesla is already using a similar approach to augment its real world data set This also doesn't even consider the fact that much of the auto industry using Nvidia "tools" will be forced to pay 50%+ margins just to buy the toolkit and will be locked in to Nvidia's system Nor does this touch on legacy auto needing to hire top ML engineering talent to actually put these tools to work The path to autonomy will be real world data as the foundation and simulations/synthetic data as a supplement. There is no path to autonomy with synthetic data alone. More to come later $TSLA As Elon said earlier this year, "it's remarkable how quickly we run out of human-created data. Reality itself and synthetic data ftw" "What you are seeing here is purely generated video sequences - given the past videos the network predicts some sample from the future, hopefully the most likely sample. It is being predicted not just for one camera, but it predicts for all 8 cameras around the car jointly" - Ashok

Dillon Loomis

60,753 Aufrufe • vor 1 Jahr

🚨 $TSLA AI VP ASHOK ELLUSWAMY SAYS CURSE OF DIMENSIONALITY MAKES SELF-DRIVING AN EXTREME COMPRESSION CHALLENGE • Tesla cars stream ~2 billion tokens from 8 high-res cameras + history buffer every few seconds, but must compress it all down to just 2 actions (steering + accel/brake) • The real difficulty is learning the correct causal reasons without spurious correlations (e.g., brake because of turn signal, not a moving tree branch) • Tesla fleet generates ~500 years of driving data daily, but most is boring → they proactively identify and collect only rare/interesting events (school buses, skids, hazards) • Large fleet allows fast capture of extremely rare scenarios that are hard or unsafe to stage • With enough rare data, the system achieves proactive safety and early prediction (e.g., braking seconds before a potential crash by reading abnormal vehicle behavior) • Bottom line: self-driving needs massive high-quality edge-case data for real predictive intelligence in real time

🚨 $TSLA AI VP ASHOK ELLUSWAMY SAYS CURSE OF DIMENSIONALITY MAKES SELF-DRIVING AN EXTREME COMPRESSION CHALLENGE • Tesla cars stream ~2 billion tokens from 8 high-res cameras + history buffer every few seconds, but must compress it all down to just 2 actions (steering + accel/brake) • The real difficulty is learning the correct causal reasons without spurious correlations (e.g., brake because of turn signal, not a moving tree branch) • Tesla fleet generates ~500 years of driving data daily, but most is boring → they proactively identify and collect only rare/interesting events (school buses, skids, hazards) • Large fleet allows fast capture of extremely rare scenarios that are hard or unsafe to stage • With enough rare data, the system achieves proactive safety and early prediction (e.g., braking seconds before a potential crash by reading abnormal vehicle behavior) • Bottom line: self-driving needs massive high-quality edge-case data for real predictive intelligence in real time

Tsla Archive

79,757 Aufrufe • vor 5 Monaten

LIVE NOW - AI on Ethereum: 8004, x402 and the Botconomy AI agents are moving onchain, so Ethereum needs new rails for identity, reputation, and payments. We had Davide Crapis and Austin Griffith to break down: ERC-8004: “AI passports” for agents (identity + reputation + validation) Why reputation + verification matters when bots become the main onchain users x402: pay-per-call payments for agent-to-agent services What “agent commerce” looks like when bots can discover → trust → pay The Clawdbot/OpenClaw moment, and why the next UX is talking to your wallet Watch now. -------------- TIMESTAMPS 0:00 “AI is the new UI,” and you’ll be talking to your wallet 4:09 Is there an arms race for AI activity across chains? 7:32 The Clawdbot/OpenClaw moment: agents get OS-level power 13:18 Why crypto is native to agents 19:41 Austin’s setup: Claude bot, heartbeat loops, and real autonomy 22:28 The token incident: how a bot “got a treasury” 28:46 Prompt injection meets wallets 38:20 ERC-8004 explained: identity, reputation, validation 40:35 x402 + 8004: discovery + payment rails for agent commerce 48:38 8004 “scans” and early adoption reality 1:05:32 What agents could build when they can hire other agents 1:12:38 Tooling for the new builder era: Speedrun → Wingman 1:25:23 “Worst time to be a junior dev…” 1:28:15 Predictions: clean English becomes the sign of AI

LIVE NOW - AI on Ethereum: 8004, x402 and the Botconomy AI agents are moving onchain, so Ethereum needs new rails for identity, reputation, and payments. We had Davide Crapis and Austin Griffith to break down: ERC-8004: “AI passports” for agents (identity + reputation + validation) Why reputation + verification matters when bots become the main onchain users x402: pay-per-call payments for agent-to-agent services What “agent commerce” looks like when bots can discover → trust → pay The Clawdbot/OpenClaw moment, and why the next UX is talking to your wallet Watch now. -------------- TIMESTAMPS 0:00 “AI is the new UI,” and you’ll be talking to your wallet 4:09 Is there an arms race for AI activity across chains? 7:32 The Clawdbot/OpenClaw moment: agents get OS-level power 13:18 Why crypto is native to agents 19:41 Austin’s setup: Claude bot, heartbeat loops, and real autonomy 22:28 The token incident: how a bot “got a treasury” 28:46 Prompt injection meets wallets 38:20 ERC-8004 explained: identity, reputation, validation 40:35 x402 + 8004: discovery + payment rails for agent commerce 48:38 8004 “scans” and early adoption reality 1:05:32 What agents could build when they can hire other agents 1:12:38 Tooling for the new builder era: Speedrun → Wingman 1:25:23 “Worst time to be a junior dev…” 1:28:15 Predictions: clean English becomes the sign of AI

Bankless

46,695 Aufrufe • vor 5 Monaten

Been optimizing my ASS OFF now that jnmartin has pushed all of his progress on his Star Fox 64 Sega Dreamcast port to a private GitHub repo to collaborate with a bunch of us DC devs before release. MAN this thing is HIGHLY optimized already... We're essentially running a high-level emulator for the N64's RSP coprocessor, doing graphics transforms, matrix math, and display list conversions not only on the main SH4 CPU, but also sitting atop a high-level OpenGL driver, in real-time. Not only that, but once again, we're doing all audio synthesis and mixing also on the main SH4 CPU, so this thing is doing a literal asston on the main CPU with what might seem like a relatively straightforward port. 99% of the time, everything runs flawlessly, but when you drop a bomb on a shitfest of enemies in a densely populated area, as with the N64 original, the FPS can dip momentarily. Here's a direct hardware capture of me testing a new SH4 optimized routine out for gainz on my Sega Dreamcast... this mofo is meant be used for one-off 3D vector transforms by a single matrix which has not been preloaded into the XMTRX FP register matrix back-bank. Rather than doing a full 4x4 load on the matrix just to do a single 4D transform via the FTRV instruction, we're simply peforming 3 3D dot products against a single 3D vector, allowing us to pipeline the loads, dot products, and store operations better than doing a load all at once followed by a transform, and we aren't wasting a lane on the FPU for a 0.0f W component!

Been optimizing my ASS OFF now that jnmartin has pushed all of his progress on his Star Fox 64 Sega Dreamcast port to a private GitHub repo to collaborate with a bunch of us DC devs before release. MAN this thing is HIGHLY optimized already... We're essentially running a high-level emulator for the N64's RSP coprocessor, doing graphics transforms, matrix math, and display list conversions not only on the main SH4 CPU, but also sitting atop a high-level OpenGL driver, in real-time. Not only that, but once again, we're doing all audio synthesis and mixing also on the main SH4 CPU, so this thing is doing a literal asston on the main CPU with what might seem like a relatively straightforward port. 99% of the time, everything runs flawlessly, but when you drop a bomb on a shitfest of enemies in a densely populated area, as with the N64 original, the FPS can dip momentarily. Here's a direct hardware capture of me testing a new SH4 optimized routine out for gainz on my Sega Dreamcast... this mofo is meant be used for one-off 3D vector transforms by a single matrix which has not been preloaded into the XMTRX FP register matrix back-bank. Rather than doing a full 4x4 load on the matrix just to do a single 4D transform via the FTRV instruction, we're simply peforming 3 3D dot products against a single 3D vector, allowing us to pipeline the loads, dot products, and store operations better than doing a load all at once followed by a transform, and we aren't wasting a lane on the FPU for a 0.0f W component!

Falco Girgis

36,911 Aufrufe • vor 8 Monaten

Perfect 1v1 Drill For Finding Your LOCKDOWN Defenders! Coach LoGalbo takes you through a drill that is perfect for finding your lockdown defenders that you can rely on when the lights turn on. As Nick LoGalbo says, it's a great way to find your "Rambo" which is a defender that you can rely on to defend your opponent’s best ball handler in the full court. And, an added offensive benefit to this drill, is that it allows for offensive players to work on their ball handling in the face of defensive pressure! This 1 on 1 Contain Drill is from Nick LoGalbo's Outer Third Defense (No Middle Defense). That's why the goal of this basketball drill is to force the opponent's primary ball handlers to the outer thirds of the basketball court (and out of the middle!) as soon as they catch the ball. By doing this, it allows the defense to dictate the entry pass to one side of the floor. By keeping the ball on the outer third of the basketball court, you can overload the backside of the defense which helps control the opponents passing lanes, deters easy passes and helps eliminate the dribble drive. You can incorporate this drill every other day in practice to re-emphasize to your defenders the importance of keeping the ball handlers out of the middle of the basketball court. Instructions and Keys to the 1 on 1 Contain Drill: - Give the ball handler a cushion. This makes it easier to contain them in the outer thirds. - Chest and contest a crossover to the middle. You do not turn your body. You must beat the offensive player to the spot and turn them back to the outer thirds. - If you get beat, you sprint ahead of the ball. You do not continue shuffling. - Keep the ball in the outer thirds the entire length of the court. It does not stop after the offensive player gets past half court. - The goal of the drill is to simply keep the ball handler in the outer thirds. - The defense should focus on beating the ball to the spot and chest to contest defense. - As soon as the first group gets to halfcourt, the second group begins. - After a group makes it all the way to the opposite baseline, they switch positions and get in line to repeat the drill going the opposite direction. Coaching Tips - Once a player gets chest to contest, they must spring back to get in front of the ball. Your goal is not to be running side by side. You must get in front of the offensive player. This prevents the offensive player from getting an angle to the basket. - After running through the drill for a specified amount of time, switch directions so the offensive players must focus on using both hands as their primary dribbling hand. - It is important to mix up players rather than like positions always competing against each other, so that players get used to guarding a wide variety of positions. It forces the players to adjust defensively regarding speed and skill. - Another positive of this drill is it gives you the ability to assess who your Rambo is: players who can work the point guard up the floor.

Perfect 1v1 Drill For Finding Your LOCKDOWN Defenders! Coach LoGalbo takes you through a drill that is perfect for finding your lockdown defenders that you can rely on when the lights turn on. As Nick LoGalbo says, it's a great way to find your "Rambo" which is a defender that you can rely on to defend your opponent’s best ball handler in the full court. And, an added offensive benefit to this drill, is that it allows for offensive players to work on their ball handling in the face of defensive pressure! This 1 on 1 Contain Drill is from Nick LoGalbo's Outer Third Defense (No Middle Defense). That's why the goal of this basketball drill is to force the opponent's primary ball handlers to the outer thirds of the basketball court (and out of the middle!) as soon as they catch the ball. By doing this, it allows the defense to dictate the entry pass to one side of the floor. By keeping the ball on the outer third of the basketball court, you can overload the backside of the defense which helps control the opponents passing lanes, deters easy passes and helps eliminate the dribble drive. You can incorporate this drill every other day in practice to re-emphasize to your defenders the importance of keeping the ball handlers out of the middle of the basketball court. Instructions and Keys to the 1 on 1 Contain Drill: - Give the ball handler a cushion. This makes it easier to contain them in the outer thirds. - Chest and contest a crossover to the middle. You do not turn your body. You must beat the offensive player to the spot and turn them back to the outer thirds. - If you get beat, you sprint ahead of the ball. You do not continue shuffling. - Keep the ball in the outer thirds the entire length of the court. It does not stop after the offensive player gets past half court. - The goal of the drill is to simply keep the ball handler in the outer thirds. - The defense should focus on beating the ball to the spot and chest to contest defense. - As soon as the first group gets to halfcourt, the second group begins. - After a group makes it all the way to the opposite baseline, they switch positions and get in line to repeat the drill going the opposite direction. Coaching Tips - Once a player gets chest to contest, they must spring back to get in front of the ball. Your goal is not to be running side by side. You must get in front of the offensive player. This prevents the offensive player from getting an angle to the basket. - After running through the drill for a specified amount of time, switch directions so the offensive players must focus on using both hands as their primary dribbling hand. - It is important to mix up players rather than like positions always competing against each other, so that players get used to guarding a wide variety of positions. It forces the players to adjust defensively regarding speed and skill. - Another positive of this drill is it gives you the ability to assess who your Rambo is: players who can work the point guard up the floor.

Joe Haefner | Breakthrough Basketball

38,009 Aufrufe • vor 1 Jahr