The stock surge that followed reflected not just the earnings beat but a broader reassessment of Intel’s position in the AI ecosystem, driven in part by Nvidia CEO Jensen Huang’s public statements that CPUs will play a major role in AI data centres alongside GPUs. The optimism is real but conditional. Investors who bid up the stock are pricing in a future in which Intel’s manufacturing becomes competitive, Apple and other customers place orders, the Terafab project delivers on its ambitions, and
(We will have to revisit this claim if Musk’s Terafab comes to fruition.) And Jensen’s explanation for why even that data point exists is more candid than I expected: Nvidia could not make the multi-billion dollar equity investment that Anthropic needed to underwrite its compute commitment, AWS and Google could, and so the offtake followed the capital.
Commerce Secretary Howard Lutnick has met repeatedly over the last year with high-ranking Apple officials, including top boss Tim Cook, SpaceX chief Elon Musk and Nvidia head Jensen Huang, to try to convince them to get into business with Intel, the WSJ report said. Last month, Musk said Tesla will use Intel's next-generation 14A manufacturing process to make chips at its Terafab project, an advanced AI chip complex Musk has envisioned in Austin.
Intel Foundry has already secured ... technology at its TeraFab, which is set to be a critical component of its compute requirements. The supply chain points out that AI training chips, due to their advanced manufacturing processes and highly complex packaging, have significantly ...
Intel’s own materials on advanced packaging underline how much of modern compute capability now depends on combining multiple dies within a single package. Performance, bandwidth, thermals, and density increasingly depend on packaging and integration choices rather than on logic scaling alone. That is one reason the repeated pairing of “logic, memory, and packaging” in the Terafab discussion is significant.
It’s not a conventional factory but an integrated approach combining design, manufacturing, advanced packaging, and testing under one umbrella. However, it’s important to separate ambition from reality.
Lam Research’s technology allows for the precise etching of the through-silicon vias (TSVs) necessary for this packaging. The company’s products could meet the Terafab requirement for dense, high-performance architectures that traditional automotive chips do not require.
The TeraFab emerges when the U.S. is seeking to regain semiconductor manufacturing capacity. The CHIPS Act, investments by Intel, TSMC’s Arizona plant, Samsung’s Texas facilities, and new advanced packaging projects reflect shared concern: dependence on Asia for critical technology vital to defense, cloud, automotive, telecommunications, and AI.
Facility is designed to vertically integrate the entire chip production pipeline — design, lithography, fabrication, memory, advanced packaging, and testing — all under one roof.
Intel’s own materials on advanced packaging underline how much of modern compute capability now depends on combining multiple dies within a single package. Performance, bandwidth, thermals, and density increasingly depend on packaging and integration choices rather than on logic scaling alone. That is one reason the repeated pairing of “logic, memory, and packaging” in the Terafab discussion is significant.
Packaging is a smaller, faster, more capital-light business than fabrication. A fab requires cleanroom facilities, lithography equipment that costs hundreds of millions per unit, deep semiconductor process expertise, and construction cycles that typically run five to seven years. The $55bn Terafab figure is consistent with that scale of project. The combined facility would, on completion, include silicon fabrication, advanced packaging including panel-level packaging, printed circuit board manuf
Design happens at fabless customers, memory is made at separate fabs (Micron, Samsung, SK Hynix), and advanced packaging and test are typically done at OSAT facilities or dedicated packaging sites. Terafab's claim is that design, lithography, fabrication, memory production, advanced packaging, and testing all happen at a single site.
Last month, the company formally ... and advanced packaging for high-performance semiconductors. The agreement marks the first major customer win for Intel’s 14A process node. Industry analysts say securing a high-profile partner such as Musk ...
Intel stock jumped on reports of a preliminary agreement with Apple (AAPL) to manufacture some of the chips used in its devices, marking a breakthrough for Intel’s foundry ambitions. The speculation comes after Intel joined Tesla Inc.’s (TSLA) Terafab project in April.
The initial phase involves building the Advanced Technology Fab in Austin, a facility dedicated to producing chips for autonomous driving systems and space exploration projects. Musk emphasized the urgency of the endeavor, stating: “Either we build Terafab, or we’ll be left without chips.
SpaceX announced plans to build a US$55bn semiconductor manufacturing and advanced computing fabrication facility, known as the Terafab, in Grimes County, Texas, in partnership with Tesla.
As SpaceX outlines in their proposal, the new facility – dubbed “Terafab”, -- will be a "multi-phase, next-generation, vertically integrated semiconductor manufacturing and advanced computing fabrication facility."
Terafab would reportedly target a stated compute output of one terawatt per year, with major power and cooling constraints. The administrative document mentions SpaceX, while Tesla is also referenced in the project presentation.