Samsung Electronics' foundry division has completed the tape-out—a key design milestone—for a next-generation AI processor intended for Tesla, with manufacturing anticipated at Samsung's Texas facility using 2nm-class technology in 2027. The chip design incorporates SK Hynix memory modules, reflecting a collaboration among key semiconductor suppliers as Tesla scales AI hardware for autonomous driving and robotics applications.
Summaries like this, in your inbox every morning.
Sign up free →What happened
Samsung Electronics' foundry division has completed tape-out of a next-generation AI processor for Tesla, according to a social media post by a senior Samsung foundry engineer on 13 July. Tape-out marks the point where a chip design is ready to move into manufacturing preparation, and production is expected to occur at Samsung's Taylor, Texas facility using 2nm-class technology in 2027.
Why it matters
The milestone signals Samsung Foundry's deepening role in advancing AI infrastructure for automotive and robotics, as Tesla develops processors to support Full Self-Driving, its Optimus humanoid robot program, and AI computing infrastructure. For Samsung, success here could strengthen its competitive position in advanced contract semiconductor manufacturing.
What to watch
Samsung has not publicly confirmed the production timeline, manufacturing location, or process technology for the reported Tesla chip, so official verification remains pending. The process typically requires several months of verification and manufacturing optimization before chips can enter volume production.
On 13 July, a senior Samsung foundry engineer posted on social media that the "Tesla-Samsung A15 chip has taped out," marking a major milestone in semiconductor development. In industry parlance, tape-out signals that the chip's physical design is complete and the design is ready to transition into manufacturing preparation—a step that enables production masks to be created and the device to proceed into engineering validation and process qualification. According to industry reports, the chip will be manufactured at Samsung's Taylor, Texas fabrication facility using the company's planned 2nm-class manufacturing technology, with production anticipated in 2027; however, Samsung has not publicly confirmed any of these details. The reported chip design incorporates SK Hynix memory modules arranged around the main central compute die, illustrating a multi-partner approach to advanced AI hardware. Tesla has previously disclosed that it is developing next-generation AI processors to support its Full Self-Driving autonomous driving system, its Optimus humanoid robot program, and broader AI computing infrastructure, though the company has not publicly identified or detailed a product called "A15." The tape-out completion represents a critical step in a lengthy process: it typically precedes several months of verification and manufacturing optimization before chips can enter volume production. If confirmed, the program would further strengthen Samsung Foundry's standing in advanced contract semiconductor manufacturing as it brings next-generation process technologies into commercial production and expands operations at its Texas facility.
Samsung Foundry's completion of tape-out for Tesla's AI processor underscores the semiconductor industry's strategic bet on automotive and robotics applications. The reported use of SK Hynix memory modules integrated around a central compute die reflects a broader ecosystem collaboration: as Tesla pursues increasingly powerful custom silicon to differentiate its autonomous driving and robotics efforts, it is working with multiple manufacturing partners, and Samsung and SK Hynix have independently announced plans to invest in new fabrication capacity. The Taylor, Texas location positions Samsung to serve U.S.-based demand, aligning with industry trends favoring domestic and allied semiconductor production. Tape-out itself is not the end point; the process typically precedes several months of verification and manufacturing optimization before chips enter volume production, meaning the path from this milestone to commercial deployment remains extended.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
No comments yet. Be the first to share your thoughts!
Log in to join the discussion



Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime
1 minute a day. The AI essentials.
200+ sources · Email / LINE / Slack