正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

I think today’s $NVDA surprise factor could be memory driven price hikes expected in early 2027 pulling Vera Rubin orders forward and accelerating the next AI infrastructure upgrade cycle. That demand would hit just as Rubin ramps with higher rack density and better energy efficiency helping hyperscalers squeeze more... show more

Shay Boloor

447,890 subscribers

144,183 次观看 • 10 天前 •via X (Twitter)

科学技术财经新闻政治

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Dan Ives on CNBC said the AI Revolution could drive nearly $4T of spending over the next 3-4 years and believes the industry is still only “15-20%” through the broader infrastructure buildout cycle. He also called quantum computing a “derivative play on AI” and highlighted quantum stocks $IBM, $IONQ, $QBTS, and $RGTI as part of the next layer of the AI ecosystem. More importantly, Dan Ives said this next phase increasingly expands beyond just $NVDA into hyperscalers, power infrastructure, datacenters, and eventually quantum itself as the AI arms race between the U.S. and China continues accelerating. He also reiterated that demand for AI infrastructure is still accelerating and does not yet include physical AI or China, reinforcing his bullish view across the broader ecosystem tied to compute, power, and infrastructure.

Dan Ives on CNBC said the AI Revolution could drive nearly $4T of spending over the next 3-4 years and believes the industry is still only “15-20%” through the broader infrastructure buildout cycle. He also called quantum computing a “derivative play on AI” and highlighted quantum stocks $IBM, $IONQ, $QBTS, and $RGTI as part of the next layer of the AI ecosystem. More importantly, Dan Ives said this next phase increasingly expands beyond just $NVDA into hyperscalers, power infrastructure, datacenters, and eventually quantum itself as the AI arms race between the U.S. and China continues accelerating. He also reiterated that demand for AI infrastructure is still accelerating and does not yet include physical AI or China, reinforcing his bullish view across the broader ecosystem tied to compute, power, and infrastructure.

Sam Badawi

65,947 次观看 • 3 个月前

'I don't think it would be fair to raise prices in one part of the country to benefit another' On #BBCBreakfast Energy Secretary Ed Miliband was asked about the idea of 'zonal pricing' - where areas with lots of green energy schemes but less demand for power would pay a lower price for electricity, while regions with high demand and less clean power could pay more

'I don't think it would be fair to raise prices in one part of the country to benefit another' On #BBCBreakfast Energy Secretary Ed Miliband was asked about the idea of 'zonal pricing' - where areas with lots of green energy schemes but less demand for power would pay a lower price for electricity, while regions with high demand and less clean power could pay more

BBC Breakfast

22,838 次观看 • 1 年前

TESLA ISN'T JUST AHEAD IN AI, IT'S AN ORDER OF MAGNITUDE MORE EFFICIENT THAN THE REST OF THE WORLD "I actually think Tesla is ahead of the rest of the world in intelligence density of AI by an order of magnitude or more This is going to sound like a pretty bold statement, but I know what the intelligence efficiency of the big models are like, Grok and a bunch of the other models Tesla’s AI is, in terms of its memory efficiency, I think more than an order of magnitude better. The Tesla AI is very compute-efficient and very memory-efficient One of the metrics one should consider for any given AI model is the intelligence per gigabyte. Especially when you’re constrained on RAM, having an AI that has very high intelligence density per gigabyte for a given number of gigabytes, how much functionality can you get out of it" - Elon Musk

TESLA ISN'T JUST AHEAD IN AI, IT'S AN ORDER OF MAGNITUDE MORE EFFICIENT THAN THE REST OF THE WORLD "I actually think Tesla is ahead of the rest of the world in intelligence density of AI by an order of magnitude or more This is going to sound like a pretty bold statement, but I know what the intelligence efficiency of the big models are like, Grok and a bunch of the other models Tesla’s AI is, in terms of its memory efficiency, I think more than an order of magnitude better. The Tesla AI is very compute-efficient and very memory-efficient One of the metrics one should consider for any given AI model is the intelligence per gigabyte. Especially when you’re constrained on RAM, having an AI that has very high intelligence density per gigabyte for a given number of gigabytes, how much functionality can you get out of it" - Elon Musk

X Freeze

62,736 次观看 • 7 个月前

AI at scale is hitting a hard limit with energy. As AI infrastructure expands, power demand, not just compute architecture, is emerging as the gating constraint and pushing frontier builders toward vertically integrated deep tech bets spanning quantum discovery and fusion energy. Lenovo Data Center's Flynn Maloy and Jason Gibson, CEO and Co-founder of Quantum Formatics, discuss their collaboration to meet this challenge. They're betting on a dual-front strategy: leveraging quantum-driven simulation to accelerate material discovery while simultaneously pursuing fusion as the long-term energy unlock. Next-generation infrastructure requires breakthroughs not just in chips, but in the physical systems that power them.

AI at scale is hitting a hard limit with energy. As AI infrastructure expands, power demand, not just compute architecture, is emerging as the gating constraint and pushing frontier builders toward vertically integrated deep tech bets spanning quantum discovery and fusion energy. Lenovo Data Center's Flynn Maloy and Jason Gibson, CEO and Co-founder of Quantum Formatics, discuss their collaboration to meet this challenge. They're betting on a dual-front strategy: leveraging quantum-driven simulation to accelerate material discovery while simultaneously pursuing fusion as the long-term energy unlock. Next-generation infrastructure requires breakthroughs not just in chips, but in the physical systems that power them.

Six Five Media

224,069 次观看 • 4 个月前

Software may be starting to show early signs of life but I still think the “hard” side of AI has the cleaner lane for 2026 and probably most of 2027. We're nowhere close to done with the physical AI buildout since compute demand is still ahead of supply, power is becoming the bottleneck and every layer of AI infrastructure still needs more capacity. I own some software names but I still see this as more of an oversold bounce than a true leadership rotation away from the companies building the AI infrastructure layer. Software probably becomes a cleaner leadership group once inference costs fall, enterprise AI budgets expand and AI features start flowing through margins but that still feels more like a 2027 story to me.

Software may be starting to show early signs of life but I still think the “hard” side of AI has the cleaner lane for 2026 and probably most of 2027. We're nowhere close to done with the physical AI buildout since compute demand is still ahead of supply, power is becoming the bottleneck and every layer of AI infrastructure still needs more capacity. I own some software names but I still see this as more of an oversold bounce than a true leadership rotation away from the companies building the AI infrastructure layer. Software probably becomes a cleaner leadership group once inference costs fall, enterprise AI budgets expand and AI features start flowing through margins but that still feels more like a 2027 story to me.

Shay Boloor

122,716 次观看 • 3 个月前

🚨 JUST NOW: JENSEN HINTED ANTHROPIC’S NEXT DEAL AND ITS LIKELY $NBIS On the $NVDA earnings call, Jensen said they’ve secured compute for Anthropic across Azure, AWS and $CRWV then added there’s a “whole list of others” they are bringing online. $NBIS is an NVIDIA Cloud Partner with early access to Vera Rubin, already deploying NVL72 across the US and Europe from H2 2026. The partnerships are in place. The infrastructure is being built. Expect a deal soon. If Jensen is flagging more capacity providers coming online for Anthropic, Nebius has everything in place to be that call.

🚨 JUST NOW: JENSEN HINTED ANTHROPIC’S NEXT DEAL AND ITS LIKELY $NBIS On the $NVDA earnings call, Jensen said they’ve secured compute for Anthropic across Azure, AWS and $CRWV then added there’s a “whole list of others” they are bringing online. $NBIS is an NVIDIA Cloud Partner with early access to Vera Rubin, already deploying NVL72 across the US and Europe from H2 2026. The partnerships are in place. The infrastructure is being built. Expect a deal soon. If Jensen is flagging more capacity providers coming online for Anthropic, Nebius has everything in place to be that call.

Jordan

159,593 次观看 • 3 个月前

$AMD's heading to $5T MC LT| Lowest $/M tokens 🧵 The real reason why Institutions are FOMOing into AMD while other Semi stocks are underperforming ($NVDA $AVGO) Not Financial Advice! DYOR! Under Dr. Lisa Su’s leadership, AMD has transformed from a distant challenger into a formidable force in AI infrastructure, delivering the industry’s most compelling TCO story for high-volume inference. Her clear vision open ecosystems, aggressive annual roadmaps, rack-scale innovation, and relentless focus on tokens-per-dollar has positioned AMD’s Helios racks as the go-to solution for hyperscalers and AI natives struggling with exploding token costs, collapsing the cost down to $0.0003-$0.0005/M tokens. I will link various threads on this analysis to supply chain and wafer ratio if you are interested in understanding the full picture. In the last 3-4 months, explosive Agentic AI demand significantly increased Inference demand for Agentic AI models with 5-10 agents. If you are a listener of CNBC or Bloomberg, u should know enterprises and companies are complaining abt cost of token, and how it starts to spike up way too much to make sense. The fact that most data center today are run by $NVDA Chips, where the cost is way too high for Training or Inference. 1. Token cost Here are some quick comp, so u understand why $META OpenAI Anthropic $MSFT $AMZN Softbank $GOOGL and many more small to medium AI Natives are buying AMD CPUs and GPUs as much as they want, or pretty much AMD chips are sold out for the next 3-5 years. Inference (Cost per Million Tokens) ~$NVDA B200 / HGX: ~$0.02–$0.08 on optimized workloads (FP4/MXFP4, speculative decoding). Significant improvement over Hopper but still premium-priced. GB200 NVL72 rack-scale: $0.05–$0.25+ ~$AMD Helios Racks: $0.0003-$0.0005 per M tokens, dramatically lower than NVIDIA equivalents in owned infra. MI355X node-level: Up to 40% more tokens per dollar vs. competing solutions ( B200), driven by higher memory capacity (up to 288GB+ HBM), strong bandwidth, and lower acquisition costs. Training ~$NVDA Rubin Rack is estimated $0.7-$1.2/M Tokens ~$AMD Helios Rack is estimated $0.65-$1.0/M Tokens 2. Why Hyperscalers and AI Natives Are Choosing AMD Token consumption (especially Agentic) is outpacing even NVIDIA’s efficiency gains, making diversification mandatory for economic viability. Massive deals reflect this reality like $META, OpenAI, $MSFT, Softbank, $AMZN, Oracle, LumaAI, G42... Dr. Lisa Su’s Vision in Action: Since taking the helm, Su has driven AMD’s turnaround with disciplined execution, annual GPU cadence (MI300 → MI350 → MI400), full-stack software (ROCm 7), open ecosystems (UALink, OCP designs), and customer-centric rack-scale solutions like Helios. Her emphasis on “tokens per dollar” and TCO has turned AMD into the pragmatic choice for sustainable AI scaling. Power/Energy Efficiency: ~Helios Rack-level is estimated at 120kW-140kW with 50% more HBM4 where Inference and Training cost matter ~Rubin Rack-Level is estimated at 160kW-230kw AMD Helios shines in owned TCO, memory density, and energy flexibility at hyperscale. Cost to build 1GW data center 1GW Helios Rack full build is estimated $30-$35B 1GW Rubin Rack full build is estimated $45-$55B 3. Superior CPUs to pair with GPUs on massive scale 5-10-20GW Agentic AI. autonomous, multi-step workflows with orchestration, tool use, parallel agents, data movement, and enterprise integration has dramatically increased the importance of strong host CPUs alongside GPUs. This shifts the CPU-to-GPU ratio higher and makes balanced systems critical toward 1:1 to 5:1 as enterprises testing more than 5-10 agents. AMD EPYC Venice excels ~Leadership core density (up to 256 Zen 6 cores per socket) for running many agents in parallel, orchestration layers, and high-throughput control-plane tasks. ~Superior performance-per-core and power efficiency ( up to 2.1x higher perf/core and 2.26x better SPECpower vs. NVIDIA Grace in benchmarks). ~Tight integration in Helios: One Venice CPU + multiple MI450 GPUs per node, enabling efficient data feeding to GPUs ("zero-copy"), parallel execution, and full rack utilization for complex agentic loops. Hyperscalers (Meta, Microsoft, Amazon, Google, Softbank) and AI natives (OpenAI, Anthropic...) are adopting high-core EPYC at scale specifically for these agentic demands, as CPUs now handle a larger share of non-model work (orchestration, policy enforcement, tool calls). This complements AMD’s lower-cost GPUs for overall TCO wins. Conclusion: NVIDIA’s Vera Rubin cannot compete with a 2 years old EPYC Turin, but AMD under Dr. Lisa Su has engineered the lowest cost-per-million-tokens, highly competitive energy-efficient solutions, and superior CPU orchestration for agentic AI at scale with Helios. Dr. Su has championed this shift since at least 2023, foreseeing the rise of agentic workflows that demand far more orchestration, parallel agents, and balanced compute well before the industry fully embraced it. Her long-term vision of AI moving from simple prompts to always-on, multi-agent systems has driven AMD’s investments in high-core EPYC CPUs and integrated rack-scale solutions, perfectly positioning the company for today’s realities. Hyperscalers and AI natives effectively have no choice but to buy more AMD system for Agentic AI as leadership in economical, power-aware, high-volume internal + agentic use. However, due to supply constraints where Supply is far behind Demand, this makes multi-vendor reality along with in-house chips drive faster industry progress, lower overall costs, and better sustainability. Not Financial Advice! DYOR! Video source: Microsoft Build 2026

$AMD's heading to $5T MC LT| Lowest $/M tokens 🧵 The real reason why Institutions are FOMOing into AMD while other Semi stocks are underperforming ($NVDA $AVGO) Not Financial Advice! DYOR! Under Dr. Lisa Su’s leadership, AMD has transformed from a distant challenger into a formidable force in AI infrastructure, delivering the industry’s most compelling TCO story for high-volume inference. Her clear vision open ecosystems, aggressive annual roadmaps, rack-scale innovation, and relentless focus on tokens-per-dollar has positioned AMD’s Helios racks as the go-to solution for hyperscalers and AI natives struggling with exploding token costs, collapsing the cost down to $0.0003-$0.0005/M tokens. I will link various threads on this analysis to supply chain and wafer ratio if you are interested in understanding the full picture. In the last 3-4 months, explosive Agentic AI demand significantly increased Inference demand for Agentic AI models with 5-10 agents. If you are a listener of CNBC or Bloomberg, u should know enterprises and companies are complaining abt cost of token, and how it starts to spike up way too much to make sense. The fact that most data center today are run by $NVDA Chips, where the cost is way too high for Training or Inference. 1. Token cost Here are some quick comp, so u understand why $META OpenAI Anthropic $MSFT $AMZN Softbank $GOOGL and many more small to medium AI Natives are buying AMD CPUs and GPUs as much as they want, or pretty much AMD chips are sold out for the next 3-5 years. Inference (Cost per Million Tokens) ~$NVDA B200 / HGX: ~$0.02–$0.08 on optimized workloads (FP4/MXFP4, speculative decoding). Significant improvement over Hopper but still premium-priced. GB200 NVL72 rack-scale: $0.05–$0.25+ ~$AMD Helios Racks: $0.0003-$0.0005 per M tokens, dramatically lower than NVIDIA equivalents in owned infra. MI355X node-level: Up to 40% more tokens per dollar vs. competing solutions ( B200), driven by higher memory capacity (up to 288GB+ HBM), strong bandwidth, and lower acquisition costs. Training ~$NVDA Rubin Rack is estimated $0.7-$1.2/M Tokens ~$AMD Helios Rack is estimated $0.65-$1.0/M Tokens 2. Why Hyperscalers and AI Natives Are Choosing AMD Token consumption (especially Agentic) is outpacing even NVIDIA’s efficiency gains, making diversification mandatory for economic viability. Massive deals reflect this reality like $META, OpenAI, $MSFT, Softbank, $AMZN, Oracle, LumaAI, G42... Dr. Lisa Su’s Vision in Action: Since taking the helm, Su has driven AMD’s turnaround with disciplined execution, annual GPU cadence (MI300 → MI350 → MI400), full-stack software (ROCm 7), open ecosystems (UALink, OCP designs), and customer-centric rack-scale solutions like Helios. Her emphasis on “tokens per dollar” and TCO has turned AMD into the pragmatic choice for sustainable AI scaling. Power/Energy Efficiency: ~Helios Rack-level is estimated at 120kW-140kW with 50% more HBM4 where Inference and Training cost matter ~Rubin Rack-Level is estimated at 160kW-230kw AMD Helios shines in owned TCO, memory density, and energy flexibility at hyperscale. Cost to build 1GW data center 1GW Helios Rack full build is estimated $30-$35B 1GW Rubin Rack full build is estimated $45-$55B 3. Superior CPUs to pair with GPUs on massive scale 5-10-20GW Agentic AI. autonomous, multi-step workflows with orchestration, tool use, parallel agents, data movement, and enterprise integration has dramatically increased the importance of strong host CPUs alongside GPUs. This shifts the CPU-to-GPU ratio higher and makes balanced systems critical toward 1:1 to 5:1 as enterprises testing more than 5-10 agents. AMD EPYC Venice excels ~Leadership core density (up to 256 Zen 6 cores per socket) for running many agents in parallel, orchestration layers, and high-throughput control-plane tasks. ~Superior performance-per-core and power efficiency ( up to 2.1x higher perf/core and 2.26x better SPECpower vs. NVIDIA Grace in benchmarks). ~Tight integration in Helios: One Venice CPU + multiple MI450 GPUs per node, enabling efficient data feeding to GPUs ("zero-copy"), parallel execution, and full rack utilization for complex agentic loops. Hyperscalers (Meta, Microsoft, Amazon, Google, Softbank) and AI natives (OpenAI, Anthropic...) are adopting high-core EPYC at scale specifically for these agentic demands, as CPUs now handle a larger share of non-model work (orchestration, policy enforcement, tool calls). This complements AMD’s lower-cost GPUs for overall TCO wins. Conclusion: NVIDIA’s Vera Rubin cannot compete with a 2 years old EPYC Turin, but AMD under Dr. Lisa Su has engineered the lowest cost-per-million-tokens, highly competitive energy-efficient solutions, and superior CPU orchestration for agentic AI at scale with Helios. Dr. Su has championed this shift since at least 2023, foreseeing the rise of agentic workflows that demand far more orchestration, parallel agents, and balanced compute well before the industry fully embraced it. Her long-term vision of AI moving from simple prompts to always-on, multi-agent systems has driven AMD’s investments in high-core EPYC CPUs and integrated rack-scale solutions, perfectly positioning the company for today’s realities. Hyperscalers and AI natives effectively have no choice but to buy more AMD system for Agentic AI as leadership in economical, power-aware, high-volume internal + agentic use. However, due to supply constraints where Supply is far behind Demand, this makes multi-vendor reality along with in-house chips drive faster industry progress, lower overall costs, and better sustainability. Not Financial Advice! DYOR! Video source: Microsoft Build 2026

Mike

145,992 次观看 • 3 个月前

The CEO of the world's largest asset manager just said something that should reframe how every investor thinks about the AI trade. Larry Fink, managing $11.5 trillion at BlackRock, stood at the Milken Institute Global Conference and said four words that matter, "We just don't have enough compute." "The United States is short power. We're short compute. We're short chips. And there's going to be shortages in all three and memory, four things. I actually believe a new asset class will be buying futures of compute." Think about what that means. Fink is predicting that compute becomes a tradable commodity like oil, like grain, like natural gas where investors buy forward contracts on future capacity because the shortage is so structural and so predictable that a derivatives market will emerge to price it. That is not a minor observation from a finance executive but rather the chairman of the most powerful capital allocator on the planet telling you that compute scarcity is a multi-year, investable megatrend. The data backs him up completely. Data centers will consume 70% of all memory chips produced globally in 2026. Advanced HBM production from Samsung, SK Hynix, and Micron is sold out through 2026 and into 2027 and a single AI server consumes 10-20x more memory than a conventional workload server. DRAM supply growth is running at just 16% annually while AI infrastructure demand is growing at 80%+. The chip crunch, the power crunch, and the compute crunch are not temporary dislocations, they are structural, and they will get worse before they get better. Fink also said something the bears keep getting wrong: "There is not an AI bubble. There is the opposite. We have supply shortages. Demand is growing much faster than anyone has ever anticipated." This is why the Milk Road Pro portfolio is built the way it is, long the companies producing and supplying the constrained resources: chips, memory, compute infrastructure, and power. Check out Milk Road Pro, link below to access our full thesis and plays.

The CEO of the world's largest asset manager just said something that should reframe how every investor thinks about the AI trade. Larry Fink, managing $11.5 trillion at BlackRock, stood at the Milken Institute Global Conference and said four words that matter, "We just don't have enough compute." "The United States is short power. We're short compute. We're short chips. And there's going to be shortages in all three and memory, four things. I actually believe a new asset class will be buying futures of compute." Think about what that means. Fink is predicting that compute becomes a tradable commodity like oil, like grain, like natural gas where investors buy forward contracts on future capacity because the shortage is so structural and so predictable that a derivatives market will emerge to price it. That is not a minor observation from a finance executive but rather the chairman of the most powerful capital allocator on the planet telling you that compute scarcity is a multi-year, investable megatrend. The data backs him up completely. Data centers will consume 70% of all memory chips produced globally in 2026. Advanced HBM production from Samsung, SK Hynix, and Micron is sold out through 2026 and into 2027 and a single AI server consumes 10-20x more memory than a conventional workload server. DRAM supply growth is running at just 16% annually while AI infrastructure demand is growing at 80%+. The chip crunch, the power crunch, and the compute crunch are not temporary dislocations, they are structural, and they will get worse before they get better. Fink also said something the bears keep getting wrong: "There is not an AI bubble. There is the opposite. We have supply shortages. Demand is growing much faster than anyone has ever anticipated." This is why the Milk Road Pro portfolio is built the way it is, long the companies producing and supplying the constrained resources: chips, memory, compute infrastructure, and power. Check out Milk Road Pro, link below to access our full thesis and plays.

Milk Road AI

419,755 次观看 • 4 个月前

Elon Musk predicts AI compute in space could become cheaper than terrestrial AI far sooner than most people expect....possibly within just 2–3 years Because solar panels in orbit can generate roughly 5x more usable energy than on Earth because there’s: • No atmosphere blocking sunlight • No day-night cycle • No seasonal variation • Continuous exposure to the Sun • Far higher solar energy efficiency than on Earth Space-based solar can eventually become cheaper than terrestrial solar because you don’t need massive protective structures, heavy glass, or weather-resistant infrastructure And once launch costs fall low enough with fully reusable rockets like Starship, deploying AI clusters in orbit could become economically compelling The future of AI will scale with solar-powered compute infrastructure operating directly in space

Elon Musk predicts AI compute in space could become cheaper than terrestrial AI far sooner than most people expect....possibly within just 2–3 years Because solar panels in orbit can generate roughly 5x more usable energy than on Earth because there’s: • No atmosphere blocking sunlight • No day-night cycle • No seasonal variation • Continuous exposure to the Sun • Far higher solar energy efficiency than on Earth Space-based solar can eventually become cheaper than terrestrial solar because you don’t need massive protective structures, heavy glass, or weather-resistant infrastructure And once launch costs fall low enough with fully reusable rockets like Starship, deploying AI clusters in orbit could become economically compelling The future of AI will scale with solar-powered compute infrastructure operating directly in space

X Freeze

20,162 次观看 • 3 个月前

Blackouts and power outages could be 100 times more common in the U.S. in just five years, according to a Department of Energy report released this week. The report found that rising demand and an aging power generation infrastructure threaten U.S. capacity. The news comes as energy prices are now rising faster than inflation, and as the U.S. Energy Information Administration projects record power demand in 2025 and 2026, thanks in part to the rapid rise of AI data centers. Yet, U.S. policy just took a big turn away from the fastest growing energy production technologies: renewables like wind and solar, which are also faster to bring online than legacy power sources. To dive into how Trump's "big bill" might reshape U.S. energy supply, Tim McPhillips spoke with Alexa St. John, a climate reporter with The Associated Press.

Blackouts and power outages could be 100 times more common in the U.S. in just five years, according to a Department of Energy report released this week. The report found that rising demand and an aging power generation infrastructure threaten U.S. capacity. The news comes as energy prices are now rising faster than inflation, and as the U.S. Energy Information Administration projects record power demand in 2025 and 2026, thanks in part to the rapid rise of AI data centers. Yet, U.S. policy just took a big turn away from the fastest growing energy production technologies: renewables like wind and solar, which are also faster to bring online than legacy power sources. To dive into how Trump's "big bill" might reshape U.S. energy supply, Tim McPhillips spoke with Alexa St. John, a climate reporter with The Associated Press.

PBS News

15,828 次观看 • 1 年前

Shaun Maguire on SpaceX IPO: "Starship is the railroad to space." "In 2019, it was very predictable that Starlink would work and produce a massive amount of revenue, and we called that right." "We're now in a very similar point to where Starlink was in 2019 with AI infrastructure and orbital compute." "I think people are really underestimating how far along the orbital compute product is right now." On the $22 trillion TAM: "What do you think is the TAM for intelligence? What's the TAM for all compute infrastructure 50 years from now?" "People are underappreciating how much infrastructure is going to be built in space and how much easier it is to access energy in space." "We're at the beginning of this AI era, and we're going to double or triple the total amount of intelligence that we have access to over the next decade — so yes, I absolutely believe in the TAM, and I would even argue it's an underestimate." How much is the "railroad to space" worth in the era of AI and orbital compute? Now imagine the railroad company also owns the AI and the orbital compute. Elon Musk Shaun Maguire

Shaun Maguire on SpaceX IPO: "Starship is the railroad to space." "In 2019, it was very predictable that Starlink would work and produce a massive amount of revenue, and we called that right." "We're now in a very similar point to where Starlink was in 2019 with AI infrastructure and orbital compute." "I think people are really underestimating how far along the orbital compute product is right now." On the $22 trillion TAM: "What do you think is the TAM for intelligence? What's the TAM for all compute infrastructure 50 years from now?" "People are underappreciating how much infrastructure is going to be built in space and how much easier it is to access energy in space." "We're at the beginning of this AI era, and we're going to double or triple the total amount of intelligence that we have access to over the next decade — so yes, I absolutely believe in the TAM, and I would even argue it's an underestimate." How much is the "railroad to space" worth in the era of AI and orbital compute? Now imagine the railroad company also owns the AI and the orbital compute. Elon Musk Shaun Maguire

KanekoaTheGreat

39,709 次观看 • 2 个月前

Elon Musk explains Tesla’s new Megapod concept for scaling distributed AI compute: “We’re building a Megapod design that pairs Tesla AI4 computers with x86 systems in a box...essentially digital Optimus in a box Similar to how Megapack packages batteries, a Megapod would combine a large number of AI4 and x86 systems inside one deployable unit These boxes could be placed anywhere across the country or outside the country allowing us to scale AI compute using distributed electricity production There is power available all around the world, including at our Supercharger sites. We have something like 7 gigawatts of power across the Supercharger network, and that continues to grow We could place Megapods at many of these locations and use that distributed power for AI compute I think that could be quite a big deal” Instead of relying entirely on enormous centralized data centers, Tesla could deploy modular AI-compute infrastructure wherever sufficient power is already available

Elon Musk explains Tesla’s new Megapod concept for scaling distributed AI compute: “We’re building a Megapod design that pairs Tesla AI4 computers with x86 systems in a box...essentially digital Optimus in a box Similar to how Megapack packages batteries, a Megapod would combine a large number of AI4 and x86 systems inside one deployable unit These boxes could be placed anywhere across the country or outside the country allowing us to scale AI compute using distributed electricity production There is power available all around the world, including at our Supercharger sites. We have something like 7 gigawatts of power across the Supercharger network, and that continues to grow We could place Megapods at many of these locations and use that distributed power for AI compute I think that could be quite a big deal” Instead of relying entirely on enormous centralized data centers, Tesla could deploy modular AI-compute infrastructure wherever sufficient power is already available

X Freeze

17,025 次观看 • 1 个月前

Parasma Founder & CEO Sean Cole, who previously went viral for training human brain cells to play Doom, says that using them for compute could be far more energy-efficient than today's silicon chips. "Regardless of how efficient we make silicon, we still haven't solved the efficiency aspect of it, in terms of power efficiency." "I think human brains are extremely power efficient in comparison to silicon." "There are things that we can harness [with human brain cells] on top of other stuff like sample efficiency and continual learning."

Parasma Founder & CEO Sean Cole, who previously went viral for training human brain cells to play Doom, says that using them for compute could be far more energy-efficient than today's silicon chips. "Regardless of how efficient we make silicon, we still haven't solved the efficiency aspect of it, in terms of power efficiency." "I think human brains are extremely power efficient in comparison to silicon." "There are things that we can harness [with human brain cells] on top of other stuff like sample efficiency and continual learning."

TBPN

42,692 次观看 • 22 天前

The $1 trillion narrative around NVIDIA becomes clearer when structured against actual numbers and timelines. 1. 𝗧𝗵𝗲 𝗯𝗮𝘀𝗲𝗹𝗶𝗻𝗲 In 2020, NVIDIA generated ~$10.9B in annual revenue 2. 𝗧𝗵𝗲 𝗴𝗿𝗼𝘄𝘁𝗵 𝗰𝘂𝗿𝘃𝗲 FY2022 → ~$26.97B FY2024 → ~$60.9B FY2025 → ~$130.5B FY2026 → ~$215.9B ~𝟮𝟬𝘅 𝗴𝗿𝗼𝘄𝘁𝗵 𝗶𝗻 𝘀𝗶𝘅 𝘆𝗲𝗮𝗿𝘀 🤯 3. 𝗧𝗵𝗲 𝗱𝗲𝗺𝗮𝗻𝗱 𝗲𝘅𝗽𝗮𝗻𝘀𝗶𝗼𝗻 At GTC, Jensen Huang outlined: ~$1 trillion in cumulative demand for Blackwell and Vera Rubin through 2027 up from ~$500B projected through 2026 just one year ago Demand expectations doubled within 12 months 4. 𝗧𝗵𝗲 𝘀𝗰𝗮𝗹𝗲 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 ~$1T across two product lines ≈ ~4.6x FY2026 revenue This excludes gaming, automotive, and other segments Even ~50% conversion implies a pathway toward ~$500B annual revenue within this decade 5. 𝗧𝗵𝗲 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝘆 𝗯𝗮𝗰𝗸𝗱𝗿𝗼𝗽 Top cloud companies are expected to invest ~$700B in data center infrastructure in 2026 Long-term projections point toward ~$3T–$4T annually by 2030 6. 𝗧𝗵𝗲 𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗮𝗹 𝗹𝗼𝗼𝗽 AI adoption → rising inference demand Inference demand → infrastructure investment Infrastructure → higher capability Capability → increased usage 𝐀 𝐜𝐨𝐦𝐩𝐨𝐮𝐧𝐝𝐢𝐧𝐠 𝐜𝐲𝐜𝐥𝐞 7. 𝗧𝗵𝗲 𝘀𝗵𝗶𝗳𝘁 Compute evolves into a primary economic layer Every AI interaction consumes it Every enterprise integrating AI scales it 𝗙𝗶𝗻𝗮𝗹 𝘁𝗮𝗸𝗲𝗮𝘄𝗮𝘆: The next decade aligns growth with access to compute And the companies controlling it define the pace and limits of that growth

The $1 trillion narrative around NVIDIA becomes clearer when structured against actual numbers and timelines. 1. 𝗧𝗵𝗲 𝗯𝗮𝘀𝗲𝗹𝗶𝗻𝗲 In 2020, NVIDIA generated ~$10.9B in annual revenue 2. 𝗧𝗵𝗲 𝗴𝗿𝗼𝘄𝘁𝗵 𝗰𝘂𝗿𝘃𝗲 FY2022 → ~$26.97B FY2024 → ~$60.9B FY2025 → ~$130.5B FY2026 → ~$215.9B ~𝟮𝟬𝘅 𝗴𝗿𝗼𝘄𝘁𝗵 𝗶𝗻 𝘀𝗶𝘅 𝘆𝗲𝗮𝗿𝘀 🤯 3. 𝗧𝗵𝗲 𝗱𝗲𝗺𝗮𝗻𝗱 𝗲𝘅𝗽𝗮𝗻𝘀𝗶𝗼𝗻 At GTC, Jensen Huang outlined: ~$1 trillion in cumulative demand for Blackwell and Vera Rubin through 2027 up from ~$500B projected through 2026 just one year ago Demand expectations doubled within 12 months 4. 𝗧𝗵𝗲 𝘀𝗰𝗮𝗹𝗲 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 ~$1T across two product lines ≈ ~4.6x FY2026 revenue This excludes gaming, automotive, and other segments Even ~50% conversion implies a pathway toward ~$500B annual revenue within this decade 5. 𝗧𝗵𝗲 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝘆 𝗯𝗮𝗰𝗸𝗱𝗿𝗼𝗽 Top cloud companies are expected to invest ~$700B in data center infrastructure in 2026 Long-term projections point toward ~$3T–$4T annually by 2030 6. 𝗧𝗵𝗲 𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗮𝗹 𝗹𝗼𝗼𝗽 AI adoption → rising inference demand Inference demand → infrastructure investment Infrastructure → higher capability Capability → increased usage 𝐀 𝐜𝐨𝐦𝐩𝐨𝐮𝐧𝐝𝐢𝐧𝐠 𝐜𝐲𝐜𝐥𝐞 7. 𝗧𝗵𝗲 𝘀𝗵𝗶𝗳𝘁 Compute evolves into a primary economic layer Every AI interaction consumes it Every enterprise integrating AI scales it 𝗙𝗶𝗻𝗮𝗹 𝘁𝗮𝗸𝗲𝗮𝘄𝗮𝘆: The next decade aligns growth with access to compute And the companies controlling it define the pace and limits of that growth

Jaymin Shah

306,435 次观看 • 5 个月前

“When we start consuming 4 to 5 quintillion tokens per year, 70% of the tokens are going to be generated by agents,” DigitalOcean CEO Paddy Srinivasan said. AI token demand is projected to surge by 2030, with agents expected to be one of the biggest drivers, according to Srinivasan. That shift means agentic workloads could become “multiple orders of magnitude bigger” than human-driven workflows, with agents requiring “a significant amount of compute: GPUs, CPUs, databases, memory, and cloud primitives all in one cloud.” Supported by DigitalOcean.

“When we start consuming 4 to 5 quintillion tokens per year, 70% of the tokens are going to be generated by agents,” DigitalOcean CEO Paddy Srinivasan said. AI token demand is projected to surge by 2030, with agents expected to be one of the biggest drivers, according to Srinivasan. That shift means agentic workloads could become “multiple orders of magnitude bigger” than human-driven workflows, with agents requiring “a significant amount of compute: GPUs, CPUs, databases, memory, and cloud primitives all in one cloud.” Supported by DigitalOcean.

The Information

16,907 次观看 • 2 个月前

Nvidia's Jensen Huang on Fox Business: * Asked for a single word on the rest of the year, Huang says "demand accelerating" * Says the reason is AI does productive work and the tokens labs generate are "profitable tokens," so what's holding it back now is just compute. * Next year is already sold out and supply-constrained at 70% growth. * "Every single [chip] that we can make has already been sold." * Customers commit a year-plus ahead because "you're not buying a computer, you're building a factory," at $50-60 billion each. * With memory prices way up, Nvidia reset margin guidance this quarter to "rip the band-aid off." * Margins fall from 75% to 72 - 73% next year after absorbing memory cost increases and repricing, to clear what he called "a cloud on our stock." * Next year's Vera Rubin will ramp even faster than Blackwell did, which Huang calls "the fastest product launch we've ever had in history." * Bottlenecks "literally everywhere," the biggest buildout in history. * Constraints span TSMC CoWoS packaging, wafers, memory, networking, silicon photonics, power generation, and "high-quality land that's powered," plus a skilled-labor shortage in the US * Says scarcity is not a bad thing. This quarter's 100% growth and next year's 70% are both supply-constrained, which can be healthy because it "causes the best technologies to emerge and rise" and makes everybody sharper. * AWS alone is set to add 2 million GPUs across 2027 and 2028. Half of revenue comes from about six hyperscalers, much of it rented onward to "Nvidia developers all over the world" * Says on hyperscalers, the Nvidia platform is "the most rentable" and "captures the highest pricing" of anything hyperscalers own. * After Bill Gates suggested taxing tokens over AI job losses, Huang says: "I see things very differently than he does." * Argues AI is "a net job creator at a scale that we've never seen" because productive companies hire more people not lay people off Plus, more. See video link in reply. Bookmark & watch: _____ More on why Nvidia needed to "rip the band-aid" due to memory makers:

Nvidia's Jensen Huang on Fox Business: * Asked for a single word on the rest of the year, Huang says "demand accelerating" * Says the reason is AI does productive work and the tokens labs generate are "profitable tokens," so what's holding it back now is just compute. * Next year is already sold out and supply-constrained at 70% growth. * "Every single [chip] that we can make has already been sold." * Customers commit a year-plus ahead because "you're not buying a computer, you're building a factory," at $50-60 billion each. * With memory prices way up, Nvidia reset margin guidance this quarter to "rip the band-aid off." * Margins fall from 75% to 72 - 73% next year after absorbing memory cost increases and repricing, to clear what he called "a cloud on our stock." * Next year's Vera Rubin will ramp even faster than Blackwell did, which Huang calls "the fastest product launch we've ever had in history." * Bottlenecks "literally everywhere," the biggest buildout in history. * Constraints span TSMC CoWoS packaging, wafers, memory, networking, silicon photonics, power generation, and "high-quality land that's powered," plus a skilled-labor shortage in the US * Says scarcity is not a bad thing. This quarter's 100% growth and next year's 70% are both supply-constrained, which can be healthy because it "causes the best technologies to emerge and rise" and makes everybody sharper. * AWS alone is set to add 2 million GPUs across 2027 and 2028. Half of revenue comes from about six hyperscalers, much of it rented onward to "Nvidia developers all over the world" * Says on hyperscalers, the Nvidia platform is "the most rentable" and "captures the highest pricing" of anything hyperscalers own. * After Bill Gates suggested taxing tokens over AI job losses, Huang says: "I see things very differently than he does." * Argues AI is "a net job creator at a scale that we've never seen" because productive companies hire more people not lay people off Plus, more. See video link in reply. Bookmark & watch: _____ More on why Nvidia needed to "rip the band-aid" due to memory makers:

Fireside Alpha

13,018 次观看 • 8 天前

Jensen Huang just doubled NVIDIA's demand forecast to $1 Trillion through 2027 🤯 Then spent two hours explaining why that number is conservative… Here's everything today from GTC: - NemoClaw: NVIDIA's open-source enterprise AI agent stack built around OpenClaw. Jensen called OpenClaw "the operating system for personal AI" and said every company needs a strategy for it. - Space-1: NVIDIA is putting Vera Rubin data centers in orbit. Not a concept. An actual system being designed for space deployment right now. - DLSS 5: 3D-guided neural rendering that blends raw graphics with generative AI. Jensen called it the future of real-time rendering. - AWS: Deploying 1 million+ NVIDIA GPUs starting this year. Azure was the first hyperscaler to power up Vera Rubin. - Vera Rubin: NVIDIA's next-gen AI supercomputer. 10x more performance per watt than Blackwell, 700 million tokens per second, shipping later this year. - Groq 3 LPU: First chip from NVIDIA's $20B Groq acquisition. A purpose-built inference accelerator that ships Q3. NVIDIA now owns training AND inference. -Feynman: The architecture after Rubin, coming 2028. New GPU, new LPU, new CPU. NVIDIA is on a 12-month chip cadence and the treadmill never stops. - Autonomous driving: BYD, Hyundai, Nissan, and Geely building Level 4 vehicles on NVIDIA. Uber deploying NVIDIA-powered robotaxis across 28 cities by 2028. The man doubled his demand forecast to a trillion dollars, announced data centers in space, and closed the show with a robot singing country music. This is NVIDIA's world. Everyone else is just renting compute in it.

Jensen Huang just doubled NVIDIA's demand forecast to $1 Trillion through 2027 🤯 Then spent two hours explaining why that number is conservative… Here's everything today from GTC: - NemoClaw: NVIDIA's open-source enterprise AI agent stack built around OpenClaw. Jensen called OpenClaw "the operating system for personal AI" and said every company needs a strategy for it. - Space-1: NVIDIA is putting Vera Rubin data centers in orbit. Not a concept. An actual system being designed for space deployment right now. - DLSS 5: 3D-guided neural rendering that blends raw graphics with generative AI. Jensen called it the future of real-time rendering. - AWS: Deploying 1 million+ NVIDIA GPUs starting this year. Azure was the first hyperscaler to power up Vera Rubin. - Vera Rubin: NVIDIA's next-gen AI supercomputer. 10x more performance per watt than Blackwell, 700 million tokens per second, shipping later this year. - Groq 3 LPU: First chip from NVIDIA's $20B Groq acquisition. A purpose-built inference accelerator that ships Q3. NVIDIA now owns training AND inference. -Feynman: The architecture after Rubin, coming 2028. New GPU, new LPU, new CPU. NVIDIA is on a 12-month chip cadence and the treadmill never stops. - Autonomous driving: BYD, Hyundai, Nissan, and Geely building Level 4 vehicles on NVIDIA. Uber deploying NVIDIA-powered robotaxis across 28 cities by 2028. The man doubled his demand forecast to a trillion dollars, announced data centers in space, and closed the show with a robot singing country music. This is NVIDIA's world. Everyone else is just renting compute in it.

Josh Kale

45,875 次观看 • 5 个月前

$This is the next big plan for SpaceX: AI Data Centers in Space. • To achieve even a small fraction of a Kardashev Type II civilization (harnessing the full energy of the Sun), AI compute will require orders of magnitude more energy than Earth can ever provide. • Earth only intercepts about 1–2 billionths of the Sun’s total energy output. • Massive-scale AI (e.g., a million times more energy than Earth could produce) can only be powered by capturing far more solar energy in space. • Space-based solar-powered AI satellites/compute clusters are therefore inevitable. • In space, sunlight is continuous (no night, no clouds, no atmosphere), so no batteries are needed. • Solar panels in space can be extremely lightweight and cheap (no glass, no storm-proof framing required). • Cooling in space is dramatically easier and simpler: just radiate heat directly into the cold vacuum — no water, no fans, no liquids, no massive cooling infrastructure. • Most of the mass/volume of current supercomputer racks (e.g., GB300) is cooling hardware; in space that largely disappears. • The cost-effectiveness of electricity and compute in space will soon be overwhelmingly better than on Earth. • Elon’s Prediction: within ~5 years (by ~2030), the lowest-cost way to run large-scale AI will be solar-powered satellites in space. • A terawatt/year of AI compute is essentially impossible on Earth with any realistic build-out of power plants. • Scaling both power generation and cooling on Earth at the required rate is physically and politically unfeasible.$

This is the next big plan for SpaceX: AI Data Centers in Space. • To achieve even a small fraction of a Kardashev Type II civilization (harnessing the full energy of the Sun), AI compute will require orders of magnitude more energy than Earth can ever provide. • Earth only intercepts about 1–2 billionths of the Sun’s total energy output. • Massive-scale AI (e.g., a million times more energy than Earth could produce) can only be powered by capturing far more solar energy in space. • Space-based solar-powered AI satellites/compute clusters are therefore inevitable. • In space, sunlight is continuous (no night, no clouds, no atmosphere), so no batteries are needed. • Solar panels in space can be extremely lightweight and cheap (no glass, no storm-proof framing required). • Cooling in space is dramatically easier and simpler: just radiate heat directly into the cold vacuum — no water, no fans, no liquids, no massive cooling infrastructure. • Most of the mass/volume of current supercomputer racks (e.g., GB300) is cooling hardware; in space that largely disappears. • The cost-effectiveness of electricity and compute in space will soon be overwhelmingly better than on Earth. • Elon’s Prediction: within ~5 years (by ~2030), the lowest-cost way to run large-scale AI will be solar-powered satellites in space. • A terawatt/year of AI compute is essentially impossible on Earth with any realistic build-out of power plants. • Scaling both power generation and cooling on Earth at the required rate is physically and politically unfeasible.

Nic Cruz Patane

49,035 次观看 • 8 个月前

BEHIND THE CAPEX BOOM: In an interview with CNBC’s Michael Santoli this week, we explore how to think about the next leg of the AI capex cycle. There’s been a world awash with liquidity for the first half of 2026, and conditions are set to tighten. But this doesn’t need to overly upset the AI buildout. What it does mean, though, is that capex will likely be more expensive. And there are a few reasons for that: 📈 Yields have been elevated and the 10-year is close to 4.8%, a level that’s tended to dent risk appetite. 💰With a higher-for-longer environment plus spreads beginning to widen, this simply increases the cost of borrowing that will support the AI buildout for companies that have increased risk profiles due to pressure on free cash flow. ⚡️The buildout will require more investment in communities. Power, as an example, where there is a real need for more spend. One consideration: Will hyperscalers need to help subsidize the grid in regions with constrained capacity? This is a building idea, top of mind for infrastructure investors. 🧾 The price tag is rising partially because the bill is getting longer. Add supply chain constraints, higher energy costs and a memory shortage and you easily see why a more than $800 billion buildout this year can turn closer to $1.5 trillion for next year. Full conversation in the link. Thank you to CNBCOvertime for having me:

BEHIND THE CAPEX BOOM: In an interview with CNBC’s Michael Santoli this week, we explore how to think about the next leg of the AI capex cycle. There’s been a world awash with liquidity for the first half of 2026, and conditions are set to tighten. But this doesn’t need to overly upset the AI buildout. What it does mean, though, is that capex will likely be more expensive. And there are a few reasons for that: 📈 Yields have been elevated and the 10-year is close to 4.8%, a level that’s tended to dent risk appetite. 💰With a higher-for-longer environment plus spreads beginning to widen, this simply increases the cost of borrowing that will support the AI buildout for companies that have increased risk profiles due to pressure on free cash flow. ⚡️The buildout will require more investment in communities. Power, as an example, where there is a real need for more spend. One consideration: Will hyperscalers need to help subsidize the grid in regions with constrained capacity? This is a building idea, top of mind for infrastructure investors. 🧾 The price tag is rising partially because the bill is getting longer. Add supply chain constraints, higher energy costs and a memory shortage and you easily see why a more than $800 billion buildout this year can turn closer to $1.5 trillion for next year. Full conversation in the link. Thank you to CNBCOvertime for having me:

Sonali Basak

15,168 次观看 • 24 天前

$MU $SKHY $DRAM Holy Shit $AMD's Lisa Su just spent two full minutes at the 2026 AI keynote raving about HBM4 capacity as the thing that separates Helios from the competition. Not compute. Memory. "When you compare Helios to the competition, we're delivering 15% more compute, 50% more HBM4 memory capacity and memory bandwidth, and 50% more scale-out bandwidth." Notice which number is bigger. "What that means is that every Helios can deliver more performance for the largest models, more capacity for longer context, and the bandwidth to scale across thousands of racks.... More than 18,000 CDNA five GPU compute units, over 4,600 Zen 6 CPU cores, and 31 terabytes of HBM4 memory, all in a single rack." "That's what it takes to run agentic AI at scale. Today, I'm excited to announce that Helios is in full production. We have shipments on track to start at the end of the third quarter and ramping into the fourth quarter in the second half of the year. I can tell you, customer demand for Helios is extremely strong, we're extremely proud of the work that we have done across the leading AI labs to adopt Helios."

$MU $SKHY $DRAM Holy Shit $AMD's Lisa Su just spent two full minutes at the 2026 AI keynote raving about HBM4 capacity as the thing that separates Helios from the competition. Not compute. Memory. "When you compare Helios to the competition, we're delivering 15% more compute, 50% more HBM4 memory capacity and memory bandwidth, and 50% more scale-out bandwidth." Notice which number is bigger. "What that means is that every Helios can deliver more performance for the largest models, more capacity for longer context, and the bandwidth to scale across thousands of racks.... More than 18,000 CDNA five GPU compute units, over 4,600 Zen 6 CPU cores, and 31 terabytes of HBM4 memory, all in a single rack." "That's what it takes to run agentic AI at scale. Today, I'm excited to announce that Helios is in full production. We have shipments on track to start at the end of the third quarter and ramping into the fourth quarter in the second half of the year. I can tell you, customer demand for Helios is extremely strong, we're extremely proud of the work that we have done across the leading AI labs to adopt Helios."

Trade Whisperer

180,433 次观看 • 1 个月前