Elon Musk's Terafab: What It Is, and the 2 Stocks That Give You Exposure Right Now
Elon Musk's Terafab: What It Is, and the 2 Stocks That Give You Exposure Right Now
Billy Duberstein, The Motley FoolSun, March 29, 2026 at 11:05 AM UTC
0
Key Points -
Elon Musk introduced his Terafab vision last week.
Musk's vision details a seemingly unfathomable demand for chips, merely from his two biggest companies.
Before investors start valuing this massive project, it's important to take a reality check.
10 stocks we like better than EchoStar ›
In a long career of conquering huge manufacturing challenges, Elon Musk might have just undertaken his biggest challenge yet. Last week, he introduced Terafab.
Terafab marks Musk's latest in a long line of audacious bets, this time homing in on semiconductor production.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
Fabricating leading-edge semiconductors is one of the most difficult manufacturing challenges in the world -- perhaps the most difficult. So why does Musk think he can pull it off? And how can investors gain exposure to this massive project?
Semiconductor wafer etched by a tool.
Image source: Getty Images.
Terafab's goal: Produce a terawatt of compute
Musk thinks he's going to need more chips. Lots more chips -- so much so that he will need to build his own massive fab to meet his needs.
Just how large are these needs? In the future, Musk envisions needing a terawatt of compute per year. To put that in perspective, Musk noted in his presentation that the world's current AI computing manufacturing capacity is just 20 gigawatts – roughly 2% of the capacity he says he will require. While leading fabs will no doubt greatly increase their AI chip capacity going forward, Musk doesn't believe it will be at a pace nearly fast enough to satisfy the demand he envisions. That's especially true given the widespread shortages of GPUs and CPUs today, and the fact that much of the world's current fab capacity is taken up by giants such as Nvidia and others.
While Tesla (NASDAQ: TSLA) is already making chips with Samsung, Musk noted in the Terafab presentation that major chip fabs don't like to expand too quickly. After all, the chip industry has been quite cyclical in the past, so chipmakers are reluctant to expand too much in any one year in case of an economic downturn. Samsung, for its part, has lagged market leader Taiwan Semiconductor Manufacturing (NYSE: TSM) in yields and efficiency. Samsung is also the world's largest memory company and will have to invest heavily in expanding both DRAM and NAND output, in addition to its logic semiconductors.
Given the intense competition for a limited supply, it's no wonder Musk is considering building his own fab.
What is Musk going to do with all those chips?
Musk's two big companies, Tesla and the soon-to-be-public SpaceX, will be the primary users of the chips.
In Musk's presentation, he outlined a strong need for chips to power Tesla's robotaxi fleet; however, the demand will be much, much greater from Tesla's upcoming Optimus robots. While Tesla's AI chip needs will be robust, Musk predicts Optimus robots will require between 10 and 100 times that amount over time. Tesla plans to release the first Optimus robots in 2027, and Musk believes these robots could eventually outnumber humans.
But Musk believes Tesla and Optimus will account for only about 100 to 200 GW, or 10% to 20% of the terafab's capacity. So the majority of the remaining 800 GW or so will be for SpaceX's efforts. Of note, SpaceX just merged with Musk's AI company, xAI. While some of the remaining chip capacity may go to xAI's AI data centers here on earth, Musk actually intends to use the vast majority of the chipmaking capacity for space.
Not only will those ambitions include AI data centers in space -- an idea that has recently received a lot of media attention -- but Musk also outlined plans to build a space station on the moon to launch rockets to other planets. Launches from the moon would be less costly, according to Musk, because the moon doesn't have the gravitational pull of Earth. And Musk's vision isn't just for enterprises; he also sees a thriving consumer space economy, contemplating tourist trips to Saturn, among other futuristic visions.
Obviously, building an entire industrial AI-driven space economy would be a massive undertaking, and that is why Terafab will be required to meet those ambitions.
2 ways to get exposure to the Terafab
Having a Terafab is tantalizing for investors. After all, TSMC, the world's largest semiconductor foundry, is valued at $1.8 trillion. So if Terafab were 50 times the size of TSMC, would Terafab's value alone be $90 trillion?
Advertisement
Of course, Terafab might not be its own standalone company. The presentation indicates it will be co-owned and managed by Tesla, currently valued at $1.4 trillion, and SpaceX, which is rumored to be valued at $1.75 trillion ahead of its upcoming IPO.
SpaceX is set to go public soon, but for those looking to jump in on SpaceX before it goes public -- especially as it will consume the majority of Terafab's theoretical output -- EchoStar (NASDAQ: SATS) currently functions as a proxy for SpaceX stock. EchoStar was previously a declining satellite cable and wireless business, but it sold SpaceX valuable spectrum last summer, partially in exchange for SpaceX stock. Today, that stake in SpaceX accounts for the majority of EchoStar's market value.
So the two ways to "invest" in Terafab are currently Tesla and EchoStar, with SpaceX another option when it goes public.
But the Terfab vision seems pretty far-fetched right now
It should be noted that Musk's vision seems a bit untethered from reality today, at least without significant technological breakthroughs.
Semiconductor manufacturing is perhaps the most challenging and capital-intensive endeavor on Earth, and it's becoming increasingly difficult as today's transistors shrink to atomic levels. Getting enough chip expertise to design, manufacture, test, and package a terawatt of compute at good yields would be a monumental undertaking. Merely assembling the chip design and manufacturing talent to accomplish this goal would be daunting, as Musk would have to poach the very best minds from all the other major fabs and chip designers.
But even beyond the talent challenges, the financial costs would also be enormous. For reference, TSMC had $117.4 billion in property, plant, and equipment at the end of 2025, and management forecasts capital spending of $52 billion to $56 billion in 2026 -- its highest-ever annual investment.
Theoretically, a Terafab producing 50 times the amount of chips TSMC makes would cost some $6 trillion in capital spending alone. Even assuming Musk finds some incredible way to produce chips twice as efficiently as the most efficient chip producer today, that would still be a $3 trillion investment. While SpaceX is rumored to be raising $75 billion in its upcoming IPO, that would account for only 2% or less of the required Terafab spending.
Needless to say, investors may want to exercise caution when factoring Terafab into these companies' valuations. There will be a lot of time and money spent between now and any sort of production or financial return.
Should you buy stock in EchoStar right now?
Before you buy stock in EchoStar, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and EchoStar wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $503,861!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,026,987!*
Now, it’s worth noting Stock Advisor’s total average return is 884% — a market-crushing outperformance compared to 179% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
See the 10 stocks »
*Stock Advisor returns as of March 29, 2026.
Billy Duberstein and/or his clients have positions in Taiwan Semiconductor Manufacturing. The Motley Fool has positions in and recommends Nvidia, Taiwan Semiconductor Manufacturing, and Tesla. The Motley Fool has a disclosure policy.
Source: “AOL Money”
