The Semiconductor theme encompasses the entire value chain of "chips"—from design (fabless) and manufacturing (foundries) to the equipment (WFE) and materials required to build them, through advanced packaging, wireless/mobile RF front-ends, and the end devices (PCs, phones) that embed them.

Semiconductors are the "new oil" of the digital economy. The transition to AI-centric computing is driving a multi-year secular growth cycle that is less dependent on traditional PC/Smartphone cycles and more on the structural buildout of AI data centers and edge AI devices. Memory and custom hyperscaler silicon are major profit engines, while mobile chipmakers diversify into automotive, IoT, and custom server compute.

Key drivers:

• Generative AI & Custom Hyperscaler Silicon: Requires massive amounts of high-performance compute (GPUs) and custom ASICs/CPUs (Broadcom, Qualcomm Dragonfly C1000) alongside specialized memory (HBM).
• Edge Computing & Mobile RF Consolidation: Shift of AI processing to local devices (AI PCs, smartphones) drives content expansion in wireless RF solutions (Skyworks, Qorvo, Qualcomm). Skyworks' $22B acquisition of Qorvo creates a $7B+ combined RF platform.
• Digital Sovereignty: Nations are subsidizing domestic chip manufacturing (CHIPS Act) to secure supply chains, leading to a "super-cycle" in Wafer Fabrication Equipment (WFE) spending. CY 2026 WFE outlook has been raised to USD 140 billion.
• Efficiency Paradox: New algorithms like TurboQuant (KV cache compression) may improve AI memory usage efficiency, but Jevons Paradox suggests higher efficiency ultimately drives greater aggregate compute demand.
• Foundry Diversification: Geopolitical demand for non-TSMC leading-edge capacity (NVIDIA, Apple, AMD seeking alternatives) is accelerating investment in at-scale challengers like Samsung.
• Device-Level AI: AI PCs and edge devices embedding models are driving a hardware refresh cycle beyond the data center.

• Compute & Connectivity Leaders: NVIDIA Corporation, AMD, Broadcom Inc., Marvell, Qualcomm Inc. (mobile + automotive + custom data center CPU), Skyworks Solutions Inc. (mobile & broad markets RF), Qorvo Inc. (RF & defense).
• Foundry/Manufacturing: Taiwan Semiconductor Manufacturing Co. (leading foundry), Samsung Electronics Co. Ltd. (integrated memory + foundry; USD 648B domestic expansion), Intel Corporation (foundry turnaround).
• Memory: Samsung Electronics Co. Ltd. (HBM4E first-to-ship), SK Hynix Inc., Micron Technology, Inc., Sandisk Corporation, Kioxia Holdings Corporation.
• WFE/Subsystems: Applied Materials, Lam Research, KLA Corporation, ASML Holding N.V., VAT Group AG, INFICON Holding AG, Comet Holding AG.
• End Devices: Dell Technologies Inc., Apple Inc. — AI PC/device refresh cycle and downstream demand for advanced silicon.

• Geopolitical Conflict & Export Controls: High concentration of manufacturing in Taiwan and tightening US-China restrictions (e.g. SAMR regulatory reviews on cross-border semiconductor M&A like SWKS/QRVO).
• Memory Supply Bottlenecks: DRAM capacity reallocation to HBM constrains consumer electronics supply chains and inflates input costs for mobile OEMs.
• Customer Concentration: Exposure to major handset OEMs (Apple modem transition impacting Qualcomm; Apple socket share uncertainty impacting Skyworks/Qorvo).

• 2026-09-01: Qualcomm Inc. unveiled FY29 non-handset revenue target of $40B (including >$15B in data center custom silicon) with Meta and Microsoft as Dragonfly anchor customers, while Skyworks Solutions Inc. advanced its $22B acquisition of Qorvo Inc. into SAMR Phase III review.
• 2026-08-26: NVIDIA Corporation reported FQ2 FY2027 revenue of $96.22B (+105.8% Y/Y) and guided FY2028 revenue growth to ~70% — calling the figure supply-constrained by memory bottlenecks.
• 2026-08-01: Samsung DS ~70% op margin Q2 2026; first to ship HBM4E samples; foundry pursuing USD 648B domestic expansion.
• 2026-05-03: WFE leaders Applied Materials, Lam Research, and KLA Corporation provided bullish outlooks through 2027 with 20%+ growth.

• AI Infrastructure
• Robotics
• Autonomous Driving
• Memory Semiconductors