techwars

🚨 BREAKING 🇨🇳 | The Chip Monopoly Is Being Challenged

China has reportedly built a prototype EUV lithography machine — the most critical and complex tool needed to manufacture advanced semiconductor chips, according to Reuters.

Until now, ASML held a near-total monopoly on EUV technology. Each machine costs around $250M and is essential for producing chips used by Nvidia, AMD, and manufactured by TSMC, Intel, and Samsung.

🔥 What’s New:

Prototype developed in Shenzhen

Successfully generated EUV light

Currently under testing

No working chips yet

2028–2030 cited as target timeline

⚡ Why It’s Huge:

Part of a 6-year state-backed program

Compared to China’s “Manhattan Project” for semiconductors

Former ASML engineers reportedly involved

Huawei coordinating across labs and suppliers

🎯 Goal:
Build advanced chips on 100% China-made machines, fully independent from U.S. and allied supply chains.

🧠 Big Picture:

If successful, this could:

Break ASML’s EUV dominance

Reshape global chip geopolitics

Impact AI, crypto mining hardware, and high-performance computing

Accelerate the tech decoupling between China and the U.S.

⏳ Still early — but the direction is clear.

The chip war just entered a new phase.

#Semiconductors #china #EUV #Aİ #TechWars

The Oracle of Omaha just made a move that signals a tectonic shift in the compute wars, and it spells BIG TROUBLE for the current $3.5 TRILLION king!
📅 November 14th Shocker: Warren Buffett's Berkshire Hathaway revealed a massive $5.1 BILLION stake in Alphabet (Google). Why now? Because the infrastructure game just changed forever.
🤯 The Triple AI Threat to NVIDIA's Empire
The Chip Killer: Days before the filing, Google launched Ironwood (TPU v7) — a proprietary AI chip designed to match GPU power... but at an estimated 75-80% lower cost.
The Math: Training a frontier AI model costs rivals like OpenAI/Microsoft/Meta $3-4 Billion. Google, using its own Ironwood TPUs, slashes this to an estimated $600-750 Million!
Google Owns the Factory. Everyone else RENTS it.
The Model Masterpiece: Weeks later, Google dropped Gemini 3, heralded as the most advanced AI model on the planet—trained without a single NVIDIA GPU. Google’s in-house silicon is driving the AI frontier.
The Great Migration Begins:
Anthropic has already committed to an order of up to 1 MILLION Google TPUs for future models.
Rumors are swirling that other major AI labs are closely eyeing the switch to Google's vastly cheaper and more available compute.
💡 Buffett Doesn't Buy Potential. He Buys Inevitability.
Warren Buffett, who famously regretted missing Google in 2004, is now buying it in 2025. This isn't a speculative tech play; it's a vote on INEVITABILITY.
Google (\text{GOOG}) now controls the cheapest, most efficient supercomputer infrastructure on Earth.
The high price and limited supply of competing GPUs created a massive bottleneck. Google's TPUs smash that bottleneck.
📈 The Takeaway for the Markets
If Google Cloud accelerates its growth by successfully transitioning rival AI labs to its low-cost TPU infrastructure, the impact is unavoidable:
NVIDIA's Pricing Power Collapses: The era of massive GPU price inflation could be over.
AI Cost Advantage: Google secures a 400%+ cost advantage in the trillion-dollar AI race.
Fact: The company that owns the cheapest, most advanced chips owns the future.
✅ Google owns the chips.
✅ Buffett owns Google.
❌ And many are still glued to the \text{NVDA} drama!
FOLLOW DXB TRADER 1 for real-time market insights and keep ahead of the curve!
Like 👍 and share 🔄 this post to help your friends understand the biggest shift in AI compute!$BNB $XRP $ASTER
#AIEconomics #TechWars #TPUVsGPU #Alphabet #NVDACollapse

2025. gada 9. oktobris — atzīmējiet šo dienu. Ritenīši ir pagriezti. Rietumi vairs neraksta noteikumus. Tagad noteikumus nosaka Pekina.
Ķīnas Tirdzniecības ministrija tieši ir izlaidusi šoku: Paziņojums 2025 Nr. 61. Saskaņā ar šo, jebkurš produkts, kas satur 0,1% vai vairāk Ķīnas izcelsmes retzemju materiālu vai saistītās tehnoloģijas — pat ja tas ir salikts ārpus Ķīnas — nonāk Ķīnas eksporta kontroles likuma jurisdikcijā.
Tas nozīmē, ka jūsu ierīces, jūsu droni, jūsu akumulatori, jūsu mikroshēmas — pat jūsu radara sistēmas — tagad ir pakļautas Ķīnas apstiprinājumam, ja tās atgriežas pie Ķīnas piegādes ķēdē.

Omahas Oracle neseko AI hype. Viņš liek uz uzņēmumu, kas ir gatavs pārtraukt hype ciklu.
Berkshire Hathaway milzīgais $5.1 miljardu ieguldījums Google nav saistīts ar meklēšanu. Tas ir klusais mājiens uz seismisko pāreju, kas notiek zem AI nozares—pāreja, kas apdraud Nvidia dominanci tās kodolā.
Šeit ir brutālā matemātika, kas izjauc GPU impēriju:
· Izmaksu plaisa: Robežu AI modeļa apmācība konkurentiem maksā $3-4 miljardi** uz Nvidia GPU. Google to dara par **$600-750 miljoniem, izmantojot savus pielāgotos TPU.