Tesla launches $20B Terafab plan
Tesla, SpaceX and xAI announced a $20 billion Terafab AI chip factory in Austin aimed at producing next‑gen processors for robots, AVs and space compute, a move to vertically integrate silicon for real‑world autonomy. The factory is pitched as the backbone for future robotics and vehicle AI stacks that require custom silicon at scale. (ryxel.ai)
Elon Musk formally unveiled Terafab at an event held inside Austin’s defunct Seaholm Power Plant on March 21, 2026. (bloomberg.com) Project materials presented by Musk set an ambition of roughly one terawatt of AI compute per year and a target process node at the 2‑nanometer class. (teslarati.com) The Terafab plan calls for co‑locating logic, memory, advanced packaging and on‑site photomask production inside the same facility to accelerate chip design-to‑fab cycles. (tomshardware.com) Capacity targets circulated in reporting include an annual output of roughly 100–200 billion AI and memory chips and a long‑term wafer‑start goal on the order of 1,000,000 wafers per month. (forbes.com) Tesla and Musk’s teams have signaled an aggressive commissioning timetable with small‑batch AI chip production aimed for late 2026 and volume ramping scheduled through 2027. (roborhythms.com) SpaceX completed an all‑stock acquisition of xAI in early February 2026, folding Musk’s AI unit into SpaceX ahead of the Terafab announcement. (techcrunch.com) Tesla first flagged the need for an in‑house fab on its January 28, 2026 earnings call, warning of a probable chip constraint in three to four years without new capacity. (bloomberg.com) Musk identified the initial Austin site as adjacent to Tesla’s Giga Texas campus in eastern Travis County while hosting the launch downtown at Seaholm. (kxan.com) Public job postings and hiring activity in mid‑March show Tesla listing roles for factory design, program management and semiconductor‑process engineering tied to an Austin fab buildout. (basenor.com) Tesla has relied on Samsung Electronics for much of its current chip supply, a dependency Musk cited when justifying the vertical‑integration push. (bloomberg.com)
