Bros, $Tesla(TSLA.US) is still falling so much today! $NVIDIA(NVDA.US) also missed the rally, really fell into a trap. All in all, $Apple Inc.(AAPL.US) stabilized at 55, $Microsoft(MSFT.US) had a quick short-term trade, $Amazon(AMZN.US) isn't even pretending, so shameless. Watch the show tonight!
" type: "Topics" locale: "en" url: "https://longbridge.com/en/topics/39504853.md" description: "🔥Elon Musk launches Terafab: Behind the 1TW computing power goal, I'm more concerned about who can really make money. When I saw the Terafab plan, my first reaction wasn't shock, but a more realistic question: Is this a feasible industrial plan, or an extreme expression of 'computing power anxiety'? Because this 1TW goal essentially means — 👉 to directly scale up the global existing computing power system by tens of times. This isn't expansion, it's rebuilding an entire industrial system. Let's first look at what it's actually doing. This isn't just building a wafer fab..." datetime: "2026-03-25T07:31:08.000Z" locales: - [en](https://longbridge.com/en/topics/39504853.md) - [zh-CN](https://longbridge.com/zh-CN/topics/39504853.md) - [zh-HK](https://longbridge.com/zh-HK/topics/39504853.md) author: "[辰逸](https://longbridge.com/en/profiles/16318663.md)" --- #Bros, $Tesla(TSLA.US) is still falling so much today! $NVIDIA(NVDA.US) also missed the rally, really fell into a trap. All in all, $Apple Inc.(AAPL.US) stabilized at 55, $Microsoft(MSFT.US) had a quick short-term trade, $Amazon(AMZN.US) isn't even pretending, so shameless. Watch the show tonight!
🔥Elon Musk launches Terafab: Behind the 1TW computing power goal, I'm more concerned about who can actually make money When I saw the Terafab plan, my first reaction wasn't shock, but a more practical question: Is this a feasible industrial plan, or just an extreme expression of "computing power anxiety"? Because the 1TW target essentially means— 👉 To scale up the global existing computing power system by dozens of times This isn't expansion; it's rebuilding an entire industrial system. First, let's see what it's actually doing. This isn't just building a wafer fab; it's attempting something more radical: Integrating logic, memory, and packaging all within the same system. In other words: 👉 Creating a "Super IDM" Redefining the way computing power is supplied, through a path outside of $Tesla(TSLA.US) $NVIDIA(NVDA.US) $Taiwan Semiconductor(TSM.US). And it's divided into two lines: One is edge inference (cars, robots) One is space and AI infrastructure Especially the space direction, accounting for as high as 80%, which is very crucial. This indicates: Terafab isn't for traditional data centers, but serves the future "distributed computing power network." But the problems are also clear. This plan almost challenges the physical limits of the entire semiconductor industry. First, the production capacity scale is unrealistic Current global computing power supply is about 20GW Target 1TW, equivalent to 50 times This means the required wafer capacity: Close to or even exceeding the global existing total Second, capital expenditure is extremely massive According to estimates, requires trillions of dollars in investment This isn't corporate-level investment; it's closer to a national-level project Third, key capabilities are missing Especially in memory and advanced packaging Tesla has accumulated experience in design, but almost no experience in manufacturing And these two areas precisely determine the performance ceiling Fourth, talent and supply chain There are very few 2nm-level engineers globally And advanced processes rely on long-term accumulation, not short-term investment So I won't simply interpret Terafab as "a competitor has emerged." It's more like: 👉 A stress test issued to the entire industry It's asking a question: If demand grows infinitely, can the existing system bear it? This also leads to a more important conclusion: What's truly certain isn't whether Terafab will succeed, but— 👉 Computing power demand far exceeds supply capacity Under this structure, I'm more focused on "certain beneficiaries" rather than the vision itself. The first category is equipment companies $ASML(ASML.US) $Applied Materials(AMAT.US) $Lam Research(LRCX.US) $KLA(KLAC.US) $TE Connectivity(TEL.US) $Teradyne(TER.US) No matter who builds the factory, these companies will get orders They are the "basic tool providers" of the entire system The second category is the existing foundry system $Taiwan Semiconductor(TSM.US) And the Samsung system The more radical Terafab is, the more it proves: How high the barriers to advanced processes are In the short term, no one can replace them The third category is the existing computing power core $NVIDIA(NVDA.US) Before Terafab truly materializes, All AI training demands still rely on the existing GPU system So my current judgment is very direct: Terafab itself is a highly uncertain long-term plan But what it brings is highly certain industry demand This is also why I value this logic more: Not who can build this factory, But who has already secured an irreplaceable position in this "computing power expansion cycle" If this plan ultimately only achieves part of its goals, Are you more inclined to think it will change the industry landscape, Or just further strengthen the advantages of existing giants? ### Related Stocks - [ASML.US](https://longbridge.com/en/quote/ASML.US.md) - [NVDA.US](https://longbridge.com/en/quote/NVDA.US.md) - [KLAC.US](https://longbridge.com/en/quote/KLAC.US.md) - [TSLA.US](https://longbridge.com/en/quote/TSLA.US.md) - [TER.US](https://longbridge.com/en/quote/TER.US.md) - [NVDL.US](https://longbridge.com/en/quote/NVDL.US.md) - [07788.HK](https://longbridge.com/en/quote/07788.HK.md) - [07388.HK](https://longbridge.com/en/quote/07388.HK.md) - [NVDY.US](https://longbridge.com/en/quote/NVDY.US.md) - [NVDD.US](https://longbridge.com/en/quote/NVDD.US.md) - [NVDX.US](https://longbridge.com/en/quote/NVDX.US.md) - [NVDQ.US](https://longbridge.com/en/quote/NVDQ.US.md) - [TSDD.US](https://longbridge.com/en/quote/TSDD.US.md) - [TSLL.US](https://longbridge.com/en/quote/TSLL.US.md) - [TSLQ.US](https://longbridge.com/en/quote/TSLQ.US.md) - [09366.HK](https://longbridge.com/en/quote/09366.HK.md) - [07766.HK](https://longbridge.com/en/quote/07766.HK.md) - [07366.HK](https://longbridge.com/en/quote/07366.HK.md) - [TSLR.US](https://longbridge.com/en/quote/TSLR.US.md) - [TSM.US](https://longbridge.com/en/quote/TSM.US.md) - [AMAT.US](https://longbridge.com/en/quote/AMAT.US.md) - [LRCX.US](https://longbridge.com/en/quote/LRCX.US.md) - [TEL.US](https://longbridge.com/en/quote/TEL.US.md) ## Comments (1) - **买了就不要乱动 · 2026-03-25T15:04:08.000Z**: Replacing TSM isn't that easy; from mass production to implementation, it's a tough challenge, and I don't want to hear about winning through corner-cutting 🤣