---
title: "Musk officially announces the construction of the largest chip factory in history: the annual production capacity target is 50 times the current global capacity, with 80% directly serving space missions"
type: "News"
locale: "en"
url: "https://longbridge.com/en/news/280042543.md"
description: "Elon Musk announced the launch of the superchip factory plan codenamed TERAFAB, with an annual production capacity target of 1 terawatt—approximately 50 times the current global capacity, of which 80% will directly serve space missions. Jointly led by Tesla, SpaceX, and xAI, this is not only the largest manufacturing plan in human history but is also seen by analysts as an industrial endorsement for SpaceX's summer IPO, aiming for a valuation of $17.5 trillion"
datetime: "2026-03-22T06:22:25.000Z"
locales:
  - [zh-CN](https://longbridge.com/zh-CN/news/280042543.md)
  - [en](https://longbridge.com/en/news/280042543.md)
  - [zh-HK](https://longbridge.com/zh-HK/news/280042543.md)
---

> Supported Languages: [简体中文](https://longbridge.com/zh-CN/news/280042543.md) | [繁體中文](https://longbridge.com/zh-HK/news/280042543.md)


# Musk officially announces the construction of the largest chip factory in history: the annual production capacity target is 50 times the current global capacity, with 80% directly serving space missions

Elon Musk announced the launch of the large-scale chip manufacturing plan codenamed TERAFAB, with an annual production capacity target of 1 terawatt (TW) of computing power chips—approximately 50 times the current global annual chip production capacity, of which about 80% will directly serve space missions. The project is jointly led by Tesla, SpaceX, and xAI, and is located in Austin, Texas, making it the largest publicly announced manufacturing plan in human history.

On March 21, Musk held a press conference in downtown Austin, attended by Texas Governor Greg Abbott. Musk stated that the current global annual chip production capacity is about 20 gigawatts (GW), which is only about 2% of his target demand; existing suppliers—including TSMC and Micron Technology—can no longer meet Tesla's demand growth in robotics, autonomous driving, and AI. **"Either build TERAFAB, or there will be no chips," Musk said. He plans to build an advanced factory near the Austin Giga Texas facility, integrating logic chips, memory chips, and advanced packaging within the same facility, equipped with photomask manufacturing equipment, forming a complete closed-loop for design, production, and testing.**

**TERAFAB is the first unified strategic project announced by Tesla, SpaceX, and xAI in a joint entity format, with the timing coinciding with SpaceX's plan for a large-scale IPO this summer.**

According to previous reports by Bloomberg, SpaceX is expected to raise up to $50 billion, with a valuation potentially exceeding $1.75 trillion, and launching AI data centers into space is one of the core financing logics for this IPO. Musk did not provide a specific timeline for the construction of the factory and the production capacity targets.

Analysts generally hold a cautious attitude towards the engineering feasibility of the plan, pointing out that semiconductor manufacturing involves investments of tens of billions of dollars, multi-year construction cycles, and highly scarce specialized talent. However, some viewpoints suggest that **the strategic significance of TERAFAB has exceeded the chip industry itself—it integrates computing power manufacturing with space expansion into the same industrial logic, directly providing industrial-level support for SpaceX's valuation narrative.**

## Computing Power Gap: The Inevitable Logic of Self-Building

The core driver behind Musk's push for TERAFAB is his extreme judgment of future computing power demand.

According to his roadmap, **the Optimus humanoid robot alone will consume 100 to 200 GW of chip computing power; the demand for space solar AI satellite clusters will reach the terawatt level. He expects the annual production of humanoid robots to ultimately reach 1 billion to 10 billion units—current global annual car production is about 100 million vehicles, while humanoid robot production could be 10 to 100 times that of cars.**

In Musk's view, ground computing power expansion is approaching a physical ceiling. The total capacity of the U.S. power grid is about 0.5 TW, which cannot simultaneously support the overlapping demands of large-scale AI training, robot operation, and data centers; good construction sites are becoming increasingly scarce, and the NIMBY effect is also intensifying. Space, on the other hand, is entirely different—there is no day-night cycle or atmospheric attenuation, and solar energy efficiency is more than five times that of the ground, with marginal costs decreasing as scale increases. Musk predicts that within 2 to 3 years, the cost of deploying AI chips in space will be lower than deploying them on the ground The existing suppliers can no longer fill this gap. Tesla has a cooperation agreement with Samsung's Austin factory for the next-generation chips and maintains supply relationships with TSMC and Micron Technology, but Musk stated that the supply growth from these manufacturers is "far below our expectations." He pointed out that the semiconductor industry as a whole is expanding output, but the pace of expansion is still insufficient.

## Factory Structure: Full Link Closed Loop and Two Types of Chips

TERAFAB plans to complete the entire process of mask manufacturing, chip production, packaging testing, and design iteration within a single building, forming a high-speed recursive iteration closed loop.

Musk stated that, to his knowledge, this level of integration is unprecedented globally, and the iteration speed is expected to be an order of magnitude faster than existing solutions. Musk previously mentioned that the target process is 2 nanometers.

**The factory will produce two types of chips: one optimized for edge inference, primarily for the Optimus robot and Tesla vehicles; the other is a high-power chip specifically designed for space environments, which must withstand high-energy particle bombardment, radiation accumulation, and extreme temperatures. The chip design allows for operation at higher temperatures than on the ground to reduce the weight requirements of the cooling system.** In terms of demand structure, space chips will dominate—Musk expects ground computing power to remain at the level of 100 to 200 GW, while space will ultimately reach the terawatt level.

At the press conference, Musk also showcased a prototype of a 100 kW AI micro-satellite and stated that "future satellites may enter the megawatt level." In January of this year, SpaceX applied to the U.S. Federal Communications Commission (FCC) for permission to launch 1 million data center satellites into orbit.

## SpaceX IPO: The Financing Main Line of Space Computing Strategy

The release of TERAFAB coincides closely with SpaceX's IPO preparations.

According to Bloomberg, SpaceX plans to complete its IPO this summer, with a financing scale of up to $50 billion, which, if successful, would set an IPO financing record, and the company's valuation could exceed $1.75 trillion. Deploying AI data centers in space is one of the core logics for this financing, and the announcement of TERAFAB provides industrial-level support for this logic.

SpaceX completed the acquisition of xAI in February of this year, and xAI now operates as its wholly-owned subsidiary. Musk stated that the majority of the chips produced by TERAFAB are expected to be consumed by xAI, primarily for space AI model training and satellite data processing.

The collaborative relationship between Tesla and xAI has already been realized on multiple levels: Tesla sells Megapack energy storage products to xAI, some models have integrated the Grok chatbot under xAI, and in January of this year, Tesla announced a $2 billion investment in xAI and signed a framework cooperation agreement. The announcement of TERAFAB is a further upgrade of the collaborative relationship among the three companies—from product-level cooperation to jointly leading the same industrial project.

## Collaboration Among Three Companies: A Complete Link from Chips to Orbit

The strategic significance of TERAFAB lies in its integration of the dispersed capabilities of Musk's three companies into a complete industrial link Under this framework, Tesla is responsible for the chip demand side of the Optimus robot and electric vehicles, SpaceX is tasked with delivering chips and computing infrastructure into orbit with large-scale transport, and xAI operates the space AI satellite system, consuming the vast majority of chip output. Together, they form a closed loop from chip manufacturing to orbital deployment to AI computation. This is also the fundamental logic behind Musk defining the project as "a joint project of three companies" rather than a single corporate action.

Musk positions TERAFAB as "the first step for humanity to become a solar system civilization" and outlines a subsequent roadmap: **to build an electromagnetic mass driver on the Moon, using the Moon's low gravity and lack of atmosphere to directly accelerate materials to escape velocity, at which point the scale of computing power could be expanded by a thousand times, entering the petawatt (PW) range. He expressed hope to see the lunar mass driver completed within his lifetime.**

In a financial report meeting in January this year, Musk stated that **building TERAFAB is aimed at "resolving a high-probability capacity bottleneck within three to four years." If the project ultimately materializes, its impact will extend beyond the semiconductor industry itself—the global computing power supply landscape, orbital data center infrastructure, and even the engineering foundation of SpaceX's deep space missions will all change accordingly.**

## Challenges: Funding, Technology, and Talent

Analysts' reservations focus on three levels.

**In terms of funding, Morgan Stanley estimates that building a factory capable of producing 100,000 advanced logic chip wafers per month will cost about $45 billion; UBS's estimate starts at $30 billion. Baird analyst Ben Kallo directly raises the most pressing question for the market: "Where will the money come from?" Musk has not yet disclosed any financing arrangements.**

On the supply chain front, high-end extreme ultraviolet (EUV) lithography machines are almost entirely dependent on Dutch company ASML, with delivery cycles lasting 1 to 2 years, and new customers often waiting even longer. Integrating logic chips, memory chips, and advanced packaging processes within the same factory will exponentially increase system complexity; new semiconductor facilities typically cost billions of dollars and often take years to reach full production.

Regarding talent, Bernstein semiconductor analyst Stacy Rasgon stated, "Because it's Musk, I won't easily dismiss it, but I doubt this is actually easier than sending a rocket to Mars." He cited TSMC's Arizona factory as an example—this project has faced years of delays and had to airlift engineers from Taiwan to the U.S. to support capacity ramp-up. "These talents are not cabbages," Rasgon said.

It is worth noting that Musk himself has no background in semiconductor manufacturing and is perceived by outsiders to have a history of over-promising goals and timelines—during this press conference, he did not provide any specific timelines regarding the construction progress or capacity ramp-up of the factory

### Related Stocks

- [T-REX 2X Long Tesla Daily Target ETF (TSLT.US)](https://longbridge.com/en/quote/TSLT.US.md)
- [State Street® SPDR® S&P® Smcndctr ETF (XSD.US)](https://longbridge.com/en/quote/XSD.US.md)
- [T-REX 2X Inverse Tesla Daily Target ETF (TSLZ.US)](https://longbridge.com/en/quote/TSLZ.US.md)
- [Leverage Shares 2X Long TSLA Daily ETF (TSLG.US)](https://longbridge.com/en/quote/TSLG.US.md)
- [Simplify Volt TSLA Revolution ETF (TESL.US)](https://longbridge.com/en/quote/TESL.US.md)
- [iShares Semiconductor ETF (SOXX.US)](https://longbridge.com/en/quote/SOXX.US.md)
- [Tesla, Inc. (TSLA.US)](https://longbridge.com/en/quote/TSLA.US.md)
- [iShares US Aerospace & Defense ETF (ITA.US)](https://longbridge.com/en/quote/ITA.US.md)
- [GraniteShares 2x Long TSLA Daily ETF (TSLR.US)](https://longbridge.com/en/quote/TSLR.US.md)
- [GraniteShares 1.25x Long Tsla Daily ETF (TSL.US)](https://longbridge.com/en/quote/TSL.US.md)
- [Stt Strt® SPDR® S&P® Arspc & Dfnc ETF (XAR.US)](https://longbridge.com/en/quote/XAR.US.md)
- [VanEck Semiconductor ETF (SMH.US)](https://longbridge.com/en/quote/SMH.US.md)
- [Tradr 2X Short TSLA Daily ETF (TSLQ.US)](https://longbridge.com/en/quote/TSLQ.US.md)
- [Invesco Aerospace & Defense ETF (PPA.US)](https://longbridge.com/en/quote/PPA.US.md)
- [Invesco Semiconductors ETF (PSI.US)](https://longbridge.com/en/quote/PSI.US.md)
- [Direxion Daily TSLA Bull 2X Shares (TSLL.US)](https://longbridge.com/en/quote/TSLL.US.md)
- [Direxion Daily Semicondct Bull 3X ETF (SOXL.US)](https://longbridge.com/en/quote/SOXL.US.md)
- [ARK Space & Defense Innovation ETF (ARKX.US)](https://longbridge.com/en/quote/ARKX.US.md)

## Related News & Research

- [Musk Speeds Up Tesla's Next-Gen AI Chip](https://longbridge.com/en/news/279798114.md)
- [Elon Musk Found Liable By Jury For Misleading Twitter Investors In $44 Billion Deal, Faces Potential $2.5 Billion Damages](https://longbridge.com/en/news/280011779.md)
- [Elon Musk Says SpaceX Will Beat 'Everyone Else Combined' In AI, Warns Non-AI Firms Risk Extinction](https://longbridge.com/en/news/279383442.md)
- [French prosecutors suspect Musk encouraged deepfakes row to inflate X value](https://longbridge.com/en/news/280023375.md)
- [Musk says Tesla may 'tape out' next-generation AI6 chips in December](https://longbridge.com/en/news/279747158.md)