Tesla's TeraFab to Use Intel's 14A Process for AI Chip Manufacturing

Elon Musk said Tesla's TeraFab semiconductor initiative plans to use Intel's 14A fabrication process when the manufacturing node matures later this decade, speaking during Tesla's earnings call. Musk said 14A "seems like the right move" and cited respect for Intel's leadership team. The arrangement likely involves licensing Intel's process technology rather than building a node from scratch, which can take a decade. Tesla will build a $3 billion research fab at its Gigafactory Texas campus capable of processing a few thousand wafers per month. The pilot line will test new manufacturing ideas and validate production processes. SpaceX will handle high-volume chip manufacturing for TeraFab. The division of responsibilities between Tesla and SpaceX introduces governance challenges, as joint efforts require approval from both companies' boards and conflict-of-interest reviews. Musk acknowledged that "we got to figure out the rest" regarding the full manufacturing plan. Licensing process technology has precedent in the semiconductor industry. GlobalFoundries licensed Samsung's 14nm process in 2014 after failing to develop its own node, and Rapidus has licensed IBM's 2nm technology for its fabs. Whether Intel's 1.4nm-class process can be successfully ported to a TeraFab facility remains to be seen, as modern nodes are dramatically more complex than those from a decade ago.