Tesla (TSLA) has never been shy about thinking big. But what the company is set to unveil this week may be its most ambitious move yet, and most investors haven’t fully processed what it means.
On March 14, CEO Elon Musk posted on X, formerly Twitter, that the “Terafab Project launches in 7 days.” That puts the official kickoff on March 21. If Tesla pulls this off, it could reshape the entire semiconductor industry.
Here’s why Tesla shareholders should be paying close attention.
Terafab is Tesla’s plan to build a massive, vertically integrated semiconductor fabrication facility. The goal is to produce logic chips, memory, and advanced packaging all under one roof, domestically, at an enormous scale.
According to a report from Teslarati, the facility is projected to produce between 100 billion and 200 billion AI and memory chips per year, targeting 100,000 wafer starts per month. For context, that kind of output would put Tesla in the same conversation as TSMC (TSM) and Samsung, the world’s most advanced chipmakers. The facility is expected to use two-nanometer (nm) process technology, which is among the most advanced nodes currently in commercial production worldwide.
Musk first flagged the need for a chip fab at Tesla’s annual shareholder meeting last year, warning that even the best-case output from existing suppliers wouldn’t be enough to meet Tesla’s needs.
He repeated the warning on the company’s fiscal Q4 earnings call, telling investors directly, “If we don’t do the Tesla Terafab, we’re going to be limited by supplier output of chips.” The concern isn’t abstract. Musk projected that chip supply could become Tesla’s single biggest growth constraint within three to four years. That’s the window Terafab is designed to close.
“Optimus is completely useless without an AI chip,” Musk said on the earnings call. “It’s like the Tin Man from The Wizard of Oz….but even worse, at least the Tin Man could walk.”
To understand why Terafab matters so much right now, it helps to look at where Tesla stood at the end of 2025.
