🇻🇳Vietnam is rapidly becoming the new Silicon Valley of hardware: 🔹 50%+ of Samsung Galaxy phones sold globally are built here. 🔹 65% of Apple AirPods are now manufactured in Vietnamese facilities. 🔹 Google Pixel production and advanced New Product Introduction (NPI) engineering are moving in. 🔹 $4B+ Intel chip hub anchors the automated ecosystem in Ho Chi Minh City. The "Made in China" era is undergoing its biggest structural shift in 30 years. The world's most complex hardware is moving to Vietnam.

How do people still not understand that 13F filings are delayed and are not real time. Hello… ripple:native

🚨 Leopold Aschenbrenner bets $8.45 billion against AI-driven semiconductor stocks. A former researcher at OpenAI, he was fired for warning that China could steal the company’s AI models. He then turned $225 million into $5.5 billion in 12 months. His quarterly 13F filing—the mandatory disclosure by large investment funds to U.S. regulators—reveals his portfolio and exposes a massive short position across the entire AI semiconductor ecosystem. His top 10 positions: $2.04B 🔴 VanEck Semiconductor ETF $SMH [Put] $1.57B 🔴 Nvidia $NVDA [Put] $1.07B 🔴 Oracle $ORCL [Put] $1.01B 🔴 Broadcom $AVGO [Put] $969M 🔴 Advanced Micro Devices $AMD [Put] $879M 🟢 Bloom Energy $BE $724M 🟢 SanDisk $SNDK $584M 🔴 Micron $MU [Put] $556M 🟢 CoreWeave $CRWV $535M 🔴 Taiwan Semiconductor $TSM [Put] A telling case: Intel—three months ago, he held a $747 million bullish call; now, he has switched to buying puts. Meanwhile, he is doubling down on Bitcoin miners transitioning into AI data center infrastructure: Applied Digital, IREN, CleanSpark, Riot Platforms, Bitfarms, plus a new position in HIVE Digital. CleanSpark’s share count has surged by 648%. The synthetic trade: Go long on companies laying concrete and supplying power to data centers, while shorting (via puts) the chips running inside them. Bloom Energy remains one of his largest positions by value at $879 million.

Everyone is focused on GPUs and model architecture. But the real AI bottleneck now is high-quality training data. @elonmusk @StabilityAI, @AnthropicAI, and @AIatMeta should keep an eye on what Bittensor Subnet 70 is building. Data will decide the next generation of video AI.

🚨 $NVDA earnings on Wednesday could make or break this market. If results are strong, the AI trade can bounce hard again across: AI semis AI photonics memory battery storage AI power infrastructure data centers cooling

This is a fascinating rotation. Leopold Aschenbrenner, ex-OpenAI, is reportedly short Nvidia and AMD and going long Bitcoin miners for a $13.6B AI bet. Read that again. Everyone is chasing GPUs. He’s chasing POWER. The AI boom isn’t just a chip story anymore. It’s an electricity and data center story. Miners already: → Control massive power contracts → Own energy-dense infrastructure → Operate large-scale compute facilities If the bottleneck shifts from silicon to watts, the valuation narrative flips. This isn’t anti-Nvidia. It’s a regime shift trade. From chip scarcity to power scarcity. The smartest capital is moving down the stack. Most accounts are selling hype or fishing for engagement. I’m here to decode where the real money is positioning. If you’re not following yet, you’re missing structural alpha ⚡

Week of Earnings Report for $NVDA, how would the tech giant perform on Q2?

May 20 will be a critically important day for the U.S. stock market. NVIDIA will release its earnings report after the U.S. market close. The current S&P 500 is nearly supported by large-cap tech stocks tied to AI and semiconductors, which account for over 40% of its weight—and at the center of this sector stands NVIDIA. In essence, NVIDIA = the AI trade itself. The S&P 500’s rise to a record high of 7,500 reflects the market’s pricing in of the following scenario: - The AI boom is real. - Corporations will continue making massive investments in AI infrastructure. - AI-related capital expenditures will keep expanding. - NVIDIA will maintain its dominance in the market. The current market is pricing in a highly “perfect future.” Market expectations are: Revenue: ~$78 billion EPS: ~$1.77 But the key point is that the market won’t be satisfied merely with achieving $78 billion in revenue. It demands even stronger revenue growth, more bullish guidance, a clear rebound in sales to China, and continued acceleration of AI investment. In other words, NVIDIA must not only beat earnings—it must deliver results that prove “the AI bubble is still heating up.” Recent gains in NVIDIA’s stock have been fueled by approval for sales of its H200 chips to major Chinese tech firms. However, the issue is that even after Trump’s visit to China, Chinese officials have explicitly stated they have not yet purchased any H200 chips. As of now, H200 sales to China have not yet been reflected in actual revenue figures. To meet market expectations, NVIDIA needs two things: - A significant earnings beat - Clear confirmation of a recovery in China sales Right now, there are numerous sources of market anxiety: - A new Fed Chair - Oil prices exceeding $100 per barrel - Geopolitical risks in the Middle East and supply chain disruptions - Overheated AI valuations If NVIDIA delivers a strong earnings report exceeding $80 billion, the AI-driven rally could accelerate further. Conversely, if earnings meet expectations but China sales remain uncertain, the market may enter a period of consolidation while waiting for the next data point. If NVIDIA misses expectations entirely, a broad repricing across AI-related stocks could occur—compounding existing market unease and triggering significant volatility.

China just added an ENTIRE GERMANY-SIZED ELECTRICITY GRID in ONE YEAR. 497 TWh. Mostly clean. Game over for energy-constrained competitors. Look at this chart: •China’s 2025 increase: 497 TWh (340 TWh solar + 140 TWh wind) •That’s more than South Africa’s entire annual output (240 TWh) •More than Italy (270), Australia/Spain/UK (~290 each) •Roughly equals Germany’s total generation •Bigger than what France or Brazil produce in a full year In 2025 alone. Coal actually fell. Hydro/nuclear/gas ticked up slightly. But renewables crushed it. This isn’t green virtue-signaling. This is raw, strategic power projection. Why this matters for AI dominance: AI isn’t running on hype or chips alone anymore. Energy is the new oil. Training and inference at frontier scale are insanely power-hungry. A single large model can consume electricity equivalent to thousands of households. Scaling clusters to millions of GPUs? You’re talking gigawatts. China’s advantages: •Cheap, abundant, dispatchable power: They’re building at a pace the West can’t match. While the US added ~51 GW last year, China slammed in hundreds of GW. Data centers in China get electricity at fractions of US prices, spot prices in some provinces under 2.5¢/kWh.11 •Strategic infrastructure: “East Data, West Computing” routes cheap western renewable power straight to AI hubs. Overbuilt capacity means headroom for explosive AI growth. •Vertical integration: They control the solar/wind supply chain, batteries, rare earths, and now the electrons. No NIMBY delays, no endless permitting. Result? China is positioning to out-compute the competition. US/Western labs already warn about the “electron gap.” Musk and others have flagged it: the limiting factor for AI isn’t talent or capital anymore, it’s turning on the servers. China is solving that while others debate. This scale lets them train bigger models longer, run more inference cheaper, iterate faster, and deploy AI across manufacturing, surveillance, autonomous systems, and exports at costs rivals can’t touch. The Reality : While some nations virtue-signal net-zero timelines and strangle their grids with regulation, China is executing at civilizational scale. They didn’t just add capacity, they added a peer economy’s worth of low-carbon power in 12 months. This is how superpowers are built in the AI era. Energy abundance = compute abundance = technological supremacy. What happens when China’s AI clusters have effectively unlimited cheap power while others ration? The next decade of innovation won’t be evenly distributed. The grid expansion isn’t just impressive — it’s decisive.