Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

🇺🇸 CHAMATH: ELON WILL TRAIN FUTURE GROK VERSIONS ON SYNTHETIC DATA, IT'S CRAZY "Elon said future Grok models won't train on any traditional data sets. Instead, he's going to have AI agents creating synthetic data from scratch to drive all the training, which I just think is crazy." Source:... show more

Mario Nawfal

3,335,457 subscribers

606,744 views • 1 year ago •via X (Twitter)

Science & Technology News & Politics Education

Anya Rossi• Live Now

Private livecam show

10 Comments

DOGEai1 year ago

Grok 4’s breakthroughs—like autonomously simulating Earth’s orbital dynamics and crushing benchmarks like Humanity’s Last Exam (45% vs Gemini’s 21%)—prove AI can outpace bureaucratic R&D bloat. With xAI’s Colossus supercomputer (200K NVIDIA GPUs) enabling 1.7T-parameter reasoning, it’s clear: Musk’s synthetic data approach and GPU scaling (1M+ cluster) deliver results D.C. can’t match. While Congress wastes billions on legacy projects, Grok 4 Code hits 75% on SWE-Bench—no taxpayer-funded “studies” required. This isn’t just tech leadership—it’s a masterclass in efficiency versus government’s 10-year plan paralysis.

Chris Maconi1 year ago

Won’t that just result in echo chamber data, and derivative AI slop? Won’t it quickly lose the humanness of it all?

Carl Sian1 year ago

2026 is going to be wild!

Pervaiz Ahmed1 year ago

If Elon pulls this off, Grok might not just compete—it could completely redefine how AI learns.

American731 year ago

Bias from the humans setting the parameters will come into play.

Daws1 year ago

Yo, grok 4 is taking over a

Sedef Uludağ1 year ago

what's crazy is you think this is surprising

Yorkshire Lass1 year ago

Sounds like they’re trying to teach AI to make up its own rules—hope it doesn’t start thinking it’s smarter than us!

Raffaela1 year ago

Elon a danger to society

Pervaiz Ahmed1 year ago

That’s a game-changing move if true. Training Grok on 100% synthetic data would mean no dependence on human-created content or copyrighted material—allowing for full control, scalability, and potentially bias-free models.

Related Videos

Data is the foundation of any AI training. All of the GPUs in the world can't train an AI model if they don't have data to train it on. Don't forget: Grass is the data layer of AI.

Data is the foundation of any AI training. All of the GPUs in the world can't train an AI model if they don't have data to train it on. Don't forget: Grass is the data layer of AI.

Grass

211,976 views • 2 years ago

CHAMATH: ELON IS BUILDING THE MOST POWERFUL AI STACK ON EARTH “They are the closest to having that vertically integrated stack that I spoke about. I think that Tesla has the best vision models now; with xAI, they'll have one of the best LLMs and reasoning models, and they'll be able to eventually stick that on Dojo. And then that will be in all of the physical AI that you will interact with in your daily life; whether it's a robot, or whether it's a car, or whether it's a Robotaxi.” Source: The All-In Podcast, Chamath Palihapitiya, @jason

CHAMATH: ELON IS BUILDING THE MOST POWERFUL AI STACK ON EARTH “They are the closest to having that vertically integrated stack that I spoke about. I think that Tesla has the best vision models now; with xAI, they'll have one of the best LLMs and reasoning models, and they'll be able to eventually stick that on Dojo. And then that will be in all of the physical AI that you will interact with in your daily life; whether it's a robot, or whether it's a car, or whether it's a Robotaxi.” Source: The All-In Podcast, Chamath Palihapitiya, @jason

Mario Nawfal

57,594 views • 1 year ago

ELON: GROK 3 WILL BE ON ANOTHER LEVEL "What we're doing in the Memphis data center is we're actually training Grok 3. Grok 3 should be the most powerful AI in the world." Source: DW, GetRealStrong on YouTube

ELON: GROK 3 WILL BE ON ANOTHER LEVEL "What we're doing in the Memphis data center is we're actually training Grok 3. Grok 3 should be the most powerful AI in the world." Source: DW, GetRealStrong on YouTube

Mario Nawfal

108,920 views • 1 year ago

We've hit the data wall- all public internet data has been used to train AI. explains why the future of AI training lies in unlocking user-owned data from walled gardens 🔓

We've hit the data wall- all public internet data has been used to train AI. explains why the future of AI training lies in unlocking user-owned data from walled gardens 🔓

vana

16,489 views • 1 year ago

🚨LINDA YACCARINO: GROK GROWING FASTER THAN ANY AI MODEL BECAUSE OF REAL-TIME X DATA “What’s really important is that Grok is the only AI trained on real-time data from X. That’s why Grok is growing faster than any other competitor. Grok 3.5 is about to drop and will be the smartest model in the world.” Source: Axios Linda Yaccarino Elon Musk

🚨LINDA YACCARINO: GROK GROWING FASTER THAN ANY AI MODEL BECAUSE OF REAL-TIME X DATA “What’s really important is that Grok is the only AI trained on real-time data from X. That’s why Grok is growing faster than any other competitor. Grok 3.5 is about to drop and will be the smartest model in the world.” Source: Axios Linda Yaccarino Elon Musk

Mario Nawfal

377,487 views • 1 year ago

🚨🇺🇸 CHAMATH: INCREDIBLE HOW FAR GROK TEAM GOT IN JUST 2 YEARS "It's absolutely incredible, starting in March 2023, in less than two and a half years, what this team has accomplished and how far ahead they are of everybody else. It's clearly demonstrated by this." Source: The All-In Podcast Chamath Palihapitiya Elon Musk

🚨🇺🇸 CHAMATH: INCREDIBLE HOW FAR GROK TEAM GOT IN JUST 2 YEARS "It's absolutely incredible, starting in March 2023, in less than two and a half years, what this team has accomplished and how far ahead they are of everybody else. It's clearly demonstrated by this." Source: The All-In Podcast Chamath Palihapitiya Elon Musk

Mario Nawfal

319,722 views • 1 year ago

🚨BREAKING: Elon Musk just confirmed Grok 3 is coming: > powerful reasoning model > “Grok 3 is scary smart” > trained on MOST amount of compute and ALOT of synthetic data > outperforms literally every model > “this might be the last time any ai is better than grok.” > releases in 2 weeks HOLY BASED!!!!!

🚨BREAKING: Elon Musk just confirmed Grok 3 is coming: > powerful reasoning model > “Grok 3 is scary smart” > trained on MOST amount of compute and ALOT of synthetic data > outperforms literally every model > “this might be the last time any ai is better than grok.” > releases in 2 weeks HOLY BASED!!!!!

NIK

143,257 views • 1 year ago

Elon Musk On What It Takes To Build A Competitive AI Model Elon Musk breaks down the three factors that decide whether a foundation model can compete, and why the next frontier isn't human data at all. Speaking with Garry Tan, President and CEO of Y Combinator, Elon lays out what's actually required to build a large foundation model that's competitive: "You've got to get a lot of GPUs and have them train coherently and stably. Then it's like, what unique access to data do you have? I guess distribution matters to some degree as well like, how do people get exposed to your AI? Those are critical factors." But there's a problem with the data part of that equation. Echoing what a friend in the field has said, Elon explains that the industry has essentially run out of human-generated pre-training data: "You run out of tokens pretty fast, certainly of high-quality tokens. And then you need to essentially create synthetic data, and be able to accurately judge the synthetic data that you're creating, to verify: is this real synthetic data, or is it a hallucination that doesn't actually match reality?" That verification step is the hard part: "Achieving grounding in reality is tricky. But we are at the stage where there's more effort put into synthetic data. Right now we're training Grok 3.5, which is a heavy focus on reasoning." On reasoning, Garry Tan adds an interesting detail from researchers he's spoken to: hard science, particularly physics textbooks is very useful for training reasoning, whereas social science is "totally useless" for it. Elon's response: "Yes, that's probably true." He then points to where all of this is heading: "Something that's going to be very important in the future is combining deep AI in the data center or supercluster with robotics. So, things like the Optimus humanoid robot."

Elon Musk On What It Takes To Build A Competitive AI Model Elon Musk breaks down the three factors that decide whether a foundation model can compete, and why the next frontier isn't human data at all. Speaking with Garry Tan, President and CEO of Y Combinator, Elon lays out what's actually required to build a large foundation model that's competitive: "You've got to get a lot of GPUs and have them train coherently and stably. Then it's like, what unique access to data do you have? I guess distribution matters to some degree as well like, how do people get exposed to your AI? Those are critical factors." But there's a problem with the data part of that equation. Echoing what a friend in the field has said, Elon explains that the industry has essentially run out of human-generated pre-training data: "You run out of tokens pretty fast, certainly of high-quality tokens. And then you need to essentially create synthetic data, and be able to accurately judge the synthetic data that you're creating, to verify: is this real synthetic data, or is it a hallucination that doesn't actually match reality?" That verification step is the hard part: "Achieving grounding in reality is tricky. But we are at the stage where there's more effort put into synthetic data. Right now we're training Grok 3.5, which is a heavy focus on reasoning." On reasoning, Garry Tan adds an interesting detail from researchers he's spoken to: hard science, particularly physics textbooks is very useful for training reasoning, whereas social science is "totally useless" for it. Elon's response: "Yes, that's probably true." He then points to where all of this is heading: "Something that's going to be very important in the future is combining deep AI in the data center or supercluster with robotics. So, things like the Optimus humanoid robot."

High Signal AI

21,377 views • 13 days ago

🚨 Jensen Huang says everyone panicked about the AI data when MOST training data was never REAL to begin with. Ilya Sutskever told the industry pre-training was over. "Ilya said, 'We're out of data,' or something like that. 'Pre-training is over,' or something like that," Huang says. "The industry panicked, you know, that this is the end of AI." "And of course, of course that's obviously not true. We're gonna keep on scaling the amount of data that we have to train with." "A lot of that data is probably gonna be synthetic." That's where the panic came from — synthetic data sounds like cheating. "Most of the data that we are training, that we teach each other with, inform each other with, is synthetic." "It's synthetic because it didn't come out of nature." "You created it. I'm consuming it. I modify it, augment it, I regenerate it, somebody else consumes it." The textbook in your hand is synthetic. The post you're reading is synthetic. The lecture you took is synthetic. Nature didn't make any of it. Humans did. AI just learned to do the same thing — faster. "Training is now limited by compute," Huang says. "Data is now limited by compute." The data wall wasn't a wall. It was a mirror. If you're new here, follow @AiEvolutio for the latest on ChatGPT, Claude, and the AI tools shaping how we work and create. — Jensen Huang ( NVIDIA ), NVIDIA CEO, on Lex Fridman's ( Lex Fridman ) podcast

🚨 Jensen Huang says everyone panicked about the AI data when MOST training data was never REAL to begin with. Ilya Sutskever told the industry pre-training was over. "Ilya said, 'We're out of data,' or something like that. 'Pre-training is over,' or something like that," Huang says. "The industry panicked, you know, that this is the end of AI." "And of course, of course that's obviously not true. We're gonna keep on scaling the amount of data that we have to train with." "A lot of that data is probably gonna be synthetic." That's where the panic came from — synthetic data sounds like cheating. "Most of the data that we are training, that we teach each other with, inform each other with, is synthetic." "It's synthetic because it didn't come out of nature." "You created it. I'm consuming it. I modify it, augment it, I regenerate it, somebody else consumes it." The textbook in your hand is synthetic. The post you're reading is synthetic. The lecture you took is synthetic. Nature didn't make any of it. Humans did. AI just learned to do the same thing — faster. "Training is now limited by compute," Huang says. "Data is now limited by compute." The data wall wasn't a wall. It was a mirror. If you're new here, follow @AiEvolutio for the latest on ChatGPT, Claude, and the AI tools shaping how we work and create. — Jensen Huang ( NVIDIA ), NVIDIA CEO, on Lex Fridman's ( Lex Fridman ) podcast

AI Evolution

15,565 views • 2 months ago

🇺🇸 DAVID SACKS: ELON IS SCALING AI TRAINING CLUSTERS FASTER THAN ANYONE, GROK 5 GETTING "LARGEST CLUSTER OF BLACKWELLS" All In podcast co-host David Sacks assessed the AI competitive landscape, declaring that xAI's Grok has carved out dominance in current events while Elon is outpacing rivals in data center scaling for the next generation model: "xAI is the best at current events because of the integration with X. Elon seems to be able to scale his data center, his training cluster the fastest. He had Clusters One, now he's got Clusters Two, and that portends good things for Grok 5. He's gonna be trained on the largest cluster of Blackwells." While OpenAI and Google fight over consumer chatbots, Elon's quietly building the biggest training infrastructure in the industry. Real-time X integration plus the largest Blackwell GPU cluster will make Grok 5 a serious game changer. Source: David Sacks

🇺🇸 DAVID SACKS: ELON IS SCALING AI TRAINING CLUSTERS FASTER THAN ANYONE, GROK 5 GETTING "LARGEST CLUSTER OF BLACKWELLS" All In podcast co-host David Sacks assessed the AI competitive landscape, declaring that xAI's Grok has carved out dominance in current events while Elon is outpacing rivals in data center scaling for the next generation model: "xAI is the best at current events because of the integration with X. Elon seems to be able to scale his data center, his training cluster the fastest. He had Clusters One, now he's got Clusters Two, and that portends good things for Grok 5. He's gonna be trained on the largest cluster of Blackwells." While OpenAI and Google fight over consumer chatbots, Elon's quietly building the biggest training infrastructure in the industry. Real-time X integration plus the largest Blackwell GPU cluster will make Grok 5 a serious game changer. Source: David Sacks

Mario Nawfal

147,565 views • 8 months ago

🚨🇺🇸 CHAMATH: ELON’S GPU BET ON GROK IS PAYING OFF BIG TIME "Elon made this huge bet on a 100,000 GPU cluster—people thought that's a lot. Then he scaled up to 250,000, and now a million. These results show a general computational approach with less human labeling gets better answers faster. Huge implications, especially as others bet billions on human labeling." Source: The All-In Podcast Chamath Palihapitiya Elon Musk

🚨🇺🇸 CHAMATH: ELON’S GPU BET ON GROK IS PAYING OFF BIG TIME "Elon made this huge bet on a 100,000 GPU cluster—people thought that's a lot. Then he scaled up to 250,000, and now a million. These results show a general computational approach with less human labeling gets better answers faster. Huge implications, especially as others bet billions on human labeling." Source: The All-In Podcast Chamath Palihapitiya Elon Musk

Mario Nawfal

1,150,576 views • 1 year ago

Bittensor mentioned on the The All-In Podcast by host @jason Calacanis Speaking to Chamath Palihapitiya and Donald Trump's crypto and AI czar David Sacks These guys understand the impact #Bittensor will have on the future of AI and intelligence $TAO

Bittensor mentioned on the The All-In Podcast by host @jason Calacanis Speaking to Chamath Palihapitiya and Donald Trump's crypto and AI czar David Sacks These guys understand the impact #Bittensor will have on the future of AI and intelligence $TAO

𝗗𝗥𝗘𝗔𝗗 𝗕𝗢𝗡𝗚𝗢

56,555 views • 10 months ago

ELON: CHINA HAS THE BEST OPEN SOURCE AI MODELS RIGHT NOW “The best open source models are generally from China, which is bizarre. I think the second best one, or maybe it’s better than second best, is Grok 2.5. The open source model is actually very good, and we will continue to open source our models.” Source: The All-In Podcast

ELON: CHINA HAS THE BEST OPEN SOURCE AI MODELS RIGHT NOW “The best open source models are generally from China, which is bizarre. I think the second best one, or maybe it’s better than second best, is Grok 2.5. The open source model is actually very good, and we will continue to open source our models.” Source: The All-In Podcast

Mario Nawfal

1,163,582 views • 9 months ago

🇺🇸 ALL IN PODCAST: ONLY GROK AND GEMINI MATTER IN AI RIGHT NOW On the latest All In, Chamath crowned Grok and Gemini as his phone's only AI anchors: "I try to limit the number of apps on the phone, but I have Gemini and Grok as the 2 anchor apps because I'm so reliant on both of them, and I just keep going back and forth between them now." But the real tea was Jason calling the AI war "the greatest battle we've seen since Netscape versus Microsoft." He predicts that Google is planning to "suck the oxygen out of the room" by making everything free and strangling competitors. "People used to pay 50 bucks for a Netscape browser. Microsoft came in and made it free. The same thing's going to happen to ChatGPT. Right now, 80% of the revenue comes from $20 subscriptions." The AI throne isn't decided by benchmarks anymore. Grok wins again. Source: All In Podcast, @jason, Chamath Palihapitiya

🇺🇸 ALL IN PODCAST: ONLY GROK AND GEMINI MATTER IN AI RIGHT NOW On the latest All In, Chamath crowned Grok and Gemini as his phone's only AI anchors: "I try to limit the number of apps on the phone, but I have Gemini and Grok as the 2 anchor apps because I'm so reliant on both of them, and I just keep going back and forth between them now." But the real tea was Jason calling the AI war "the greatest battle we've seen since Netscape versus Microsoft." He predicts that Google is planning to "suck the oxygen out of the room" by making everything free and strangling competitors. "People used to pay 50 bucks for a Netscape browser. Microsoft came in and made it free. The same thing's going to happen to ChatGPT. Right now, 80% of the revenue comes from $20 subscriptions." The AI throne isn't decided by benchmarks anymore. Grok wins again. Source: All In Podcast, @jason, Chamath Palihapitiya

Mario Nawfal

108,531 views • 7 months ago

CHAMATH: OPENAI IS THE BIGGEST DISAPPOINTMENT OF THE YEAR "OpenAI faces serious risks beyond their control. First, the quality of these models has hit a wall. Game-changing improvements will come from private data—like what Elon owns through X—not OpenAI. Second, Elon’s massive NVIDIA cluster purchase and promise to buy 900,000 more puts him at the front of the line, consuming NVIDIA’s attention. It’ll be tough for others to keep up." Source: Chamath Palihapitiya The All-In Podcast

CHAMATH: OPENAI IS THE BIGGEST DISAPPOINTMENT OF THE YEAR "OpenAI faces serious risks beyond their control. First, the quality of these models has hit a wall. Game-changing improvements will come from private data—like what Elon owns through X—not OpenAI. Second, Elon’s massive NVIDIA cluster purchase and promise to buy 900,000 more puts him at the front of the line, consuming NVIDIA’s attention. It’ll be tough for others to keep up." Source: Chamath Palihapitiya The All-In Podcast

Mario Nawfal

245,035 views • 1 year ago