On February 11, 2026, Aehr Test Systems announced that it received the first production purchase order from a “leading hyperscale customer” for Sonoma ultra-high power test and burn-in systems, and that delivery will be made in the summer of 2026. This order is interpreted as a strong signal that a next-generation AI ASIC has moved from engineering validation to high-volume production. The main importance of the announcement is that, due to the power envelope of the package to be tested approaching 2,000W, the classic air-cooled burn-in approach is insufficient and a liquid-cooled infrastructure becomes mandatory. Aehr’s Sonoma platform, strengthened after the Incal Technology acquisition with active liquid cooling and per-device precise thermal control capability, creates a “technological moat” in this niche. In my opinion, with high probability, $GOOGL and TPU v6 (Trillium-derived). On the technical side, the association of the 2,000W threshold in the literature with TPU v6 is presented as a “power signature” that directly overlaps with Sonoma’s 2,000W capacity. On the operational side, it is described that the “leading test house in Silicon Valley” is ISE Labs (an ASE subsidiary), and that this facility’s job postings for the Google TPU program strengthen the Google connection. On the strategic side, it is argued that the statement “the current-generation processor is already in high-volume production” fits Google better than newer entrants like Microsoft, due to its multi-generation and large-scale TPU history. Financially, this development supports Aehr’s move away from EV/Silicon Carbide dependency and into the AI data center supercycle, and also supports building a more durable revenue model through consumables revenues such as BIM/sockets in addition to system sales. As a result, this win could be a reflection of the “silicon sovereignty” trend, and Aehr could benefit for multiple years from Google’s next-generation AI infrastructure ramp with its summer 2026 deliveries.
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
$AEHR
On February 11, 2026, Aehr Test Systems announced that it received the first production purchase order from a “leading hyperscale customer” for Sonoma ultra-high power test and burn-in systems, and that delivery will be made in the summer of 2026. This order is interpreted as a strong signal that a next-generation AI ASIC has moved from engineering validation to high-volume production. The main importance of the announcement is that, due to the power envelope of the package to be tested approaching 2,000W, the classic air-cooled burn-in approach is insufficient and a liquid-cooled infrastructure becomes mandatory. Aehr’s Sonoma platform, strengthened after the Incal Technology acquisition with active liquid cooling and per-device precise thermal control capability, creates a “technological moat” in this niche. In my opinion, with high probability, $GOOGL and TPU v6 (Trillium-derived).
On the technical side, the association of the 2,000W threshold in the literature with TPU v6 is presented as a “power signature” that directly overlaps with Sonoma’s 2,000W capacity. On the operational side, it is described that the “leading test house in Silicon Valley” is ISE Labs (an ASE subsidiary), and that this facility’s job postings for the Google TPU program strengthen the Google connection. On the strategic side, it is argued that the statement “the current-generation processor is already in high-volume production” fits Google better than newer entrants like Microsoft, due to its multi-generation and large-scale TPU history. Financially, this development supports Aehr’s move away from EV/Silicon Carbide dependency and into the AI data center supercycle, and also supports building a more durable revenue model through consumables revenues such as BIM/sockets in addition to system sales. As a result, this win could be a reflection of the “silicon sovereignty” trend, and Aehr could benefit for multiple years from Google’s next-generation AI infrastructure ramp with its summer 2026 deliveries.