#NVIDIA#'s #GTCBerlin# keynote spotlights the supply chain: 17 DSX POD partners incl. Delta, Gigabyte, Pegatron, Wistron, plus 6 capacity suppliers (Unimicron, IBIDEN, SEM, Amkor, KYEC, SPIL). 💡More: 🔗 $NVDA $TSM $AMKR

Taiwan's AI supply chain is going global 🌏 TSMC's US investment hits $265B, with 12 fabs planned Quanta, Wistron, Inventec: $1.3B+ into US server ops By 2030: US to host 40% of Taiwan firms' AI server capacity, Taiwan just 30% (MIC) #AI# #Taiwan#

Summary of Robust VR200 Outlook 
📈NVL72 rack TAM: 73k / 96k in 2026E/2027E • Vera Rubin ramping with 85%+ first-pass yield; further acceleration expected in 1Q27
• 2026E: 67k GB200/GB300 + 6k VR200
• 2027E: VR200 + initial VR300 (Ultra) drive most of the growth (86k units)
• SpaceX set to be largest VR200 buyer in 2027E (20k racks on 30k+ demand / ~10GW)
• Major 6 neoclouds (SpaceX + CoreWeave + tier-2s) to take 19k/34k racks (27%/35% of industry)
• CoreWeave near-term funding risk viewed as manageable Vendor dynamics
• Hon Hai winning bigger SpaceX allocation (model 40-45%) + share gains at Google/Meta
• Pegatron limited; DELL/Wistron face share pressure Recommendations
• Hon Hai (2317 TT) Buy – rising SpaceX + CSP share
 #VR200# #NVL72# #AIServer# #Neocloud# #HonHai# #Lenovo#

NVIDIA announced DSX POD, a new modular AI factory design that could reduce deployment time from six months to as little as 14 days. The new system is built around $NVDA's Vera Rubin platform and allows complete AI factories to be assembled, integrated, and tested inside modular containers before being delivered to customers. Instead of building everything on-site, NVIDIA's partners can preinstall computing systems, networking, cabling, rack-level power, and liquid cooling at manufacturing facilities. Once delivered, the containers can connect to a prepared site's infrastructure, allowing companies to start running AI workloads and generating revenue much sooner. NVIDIA is working with 17 partners to bring DSX POD AI factories to market, including: • Dell Technologies • HPE • Lenovo • Schneider Electric • Vertiv • Anduril • Armada • Bull • Crusoe • DayOne • Delta • Eaton • Foxconn (Ingrasys) • GIGABYTE • Pegatron • Quanta Cloud Technology (QCT) • Wistron NVIDIA expects these modular AI factories to become available worldwide in early 2027.

Lisa Su made a short, unannounced trip to Taiwan. The bottleneck she is chasing is not the wafer. $AMD The AI hardware crunch has moved past TSMC’s front end. Inference and agentic racks are now hitting packaging, liquid cooling, and rack-scale integration. Those steps sit with the ODMs. She is in closed-door meetings with Quanta, Wistron, and Pegatron to lock supply. AMD is also putting $10 billion into the broader ecosystem. The point is simple. Next-gen chips do not ship if the rack, the cooler, and the package are late. Front-end capacity was the 2024 story. The 2026 story is who can finish the system.

Nvidia CEO Jensen Huang says $NVDA’s all-in investment across the AI ecosystem may be over $100B. “All in, we might be like a hundred billion dollars.” That's larger than the $52.7B 🇺🇸 CHIPS Act semiconductor funding Major announced investments / commitments: • OpenAI: up to $100B tied to 10GW deployment • Anthropic: up to $10B • Intel: $5B • IREN: right to invest up to $2.1B • CoreWeave: $2B • Lumentum: $2B + multibillion-$$$ purchase commitment • Coherent: $2B + multibillion-$$$ purchase commitment • Marvell: $2B • Nebius: $2B • Synopsys: $2B • SB Energy: $1.5B • Nokia: $1B • NAVER: planned $1B • Lilly: joint investment up to $1B • SSI: investment announced, amount undisclosed There are also major purchase commitments with TSMC, Wistron, Foxconn, Amkor and SPIL. Not all of this is cash already invested. Some are multi-year, contingent or purchase commitments.