Semiconductor

Asian chip stocks declined following a drop in Nvidia’s stock, despite strong earnings. SoftBank, having recently sold its Nvidia shares, also saw a significant fall. Companies like SK Hynix, Samsung, and TSMC, key Nvidia suppliers and competitors, experienced declines. Factors contributing to the downturn include Bitcoin’s selloff, speculation regarding Fed interest rates, tightening financial conditions, and AI bubble concerns. The pullback highlights the interconnectedness and sensitivity of the semiconductor supply chain to major players like Nvidia.

Nvidia’s strong earnings, driven by AI chip demand, boosted Asian chip stocks. SK Hynix and Samsung, key memory suppliers, saw gains along with TSMC, Nvidia’s primary chip manufacturer. Renesas and Tokyo Electron, Nvidia suppliers, also rose. SoftBank, despite selling Nvidia shares, benefited through Arm and AI ventures. Nvidia’s CEO dismissed “AI bubble” concerns, citing a “new computing era,” suggesting long-term industry growth.

The semiconductor industry anticipates a memory chip shortage due to surging AI demand, diverting resources towards advanced memory solutions and impacting other sectors like consumer electronics and automotive. SMIC’s CEO noted client hesitation in placing orders for other chips. Analysts highlight under-investment in production capacity following recent downturns, exacerbating supply constraints. Memory companies are reportedly increasing prices, with potential impacts on low-end devices and consumers, who may face higher electronics prices.

Elon Musk announced Tesla may build its own “gigantic” chip fabrication plant (“Tesla terra fab”) to meet growing AI and robotics demands. Currently relying on TSMC and Samsung, Tesla could also partner with Intel. Musk envisions a capacity of 100,000 wafer starts per month, scaling to 1 million, potentially disrupting the semiconductor landscape. This signifies a major shift to vertical integration. Musk also announced the production of Cybercab in April, reinforcing Tesla’s commitment to AI and robotics as drivers of economic growth.

SoftBank Group shares tumbled over 8% amid a broader AI stock sell-off, reflecting investor concern over inflated sector valuations. This follows volatile trading for SoftBank, potentially erasing $53 billion in market cap this week, its worst performance since March 2020. Analysts attribute the decline to SoftBank being seen as an OpenAI proxy, vulnerable to shifting AI sentiment. Uncertainty around OpenAI’s partnerships and potential reliance on government funding are contributing factors. The downturn affected other Asian and U.S. tech companies, raising bubble concerns, though some experts see it as valuation fatigue rather than a collapse.

Qualcomm reported strong fiscal Q4 results, exceeding expectations with adjusted EPS of $3.00 and revenue of $11.27 billion. They forecast strong Q1 revenue and EPS, driven by handsets, automotive, and upcoming AI offerings. Facing the anticipated loss of Apple as a modem customer, Qualcomm is diversifying into PCs, VR/AR, and AI chips, directly challenging Nvidia and AMD in the data center AI space with new AI accelerator chips. Handset and automotive revenue saw significant growth, while licensing revenue declined slightly.

AMD’s Q3 results exceeded expectations with $9.25B revenue and $1.20 adjusted EPS. Revenue grew 36% YoY, driven by strong performance across key segments. Q4 revenue is projected at $9.6B, up 25% YoY, with a 54.5% gross margin. The company’s guidance excludes Instinct MI308 chip revenue to China. Strategic partnerships with OpenAI and Oracle, and Amazon’s sale of AMD shares highlight key developments. The company’s success in AI and other key areas intensifies its rivalry with Nvidia.

Samsung is investing in a 50,000 Nvidia GPU cluster, dubbed the “AI Megafactory,” to enhance chip manufacturing for mobile devices and robotics. This project signifies the escalating demand for AI acceleration and strengthens Nvidia’s position in the AI revolution. In addition to being a customer, Samsung will leverage Nvidia’s Omniverse and optimize chipmaking lithography. Samsung will also collaborate with Nvidia on HBM4 memory technology for future AI chips. These partnerships support Korea’s ambition to become a global AI leader.

SK Hynix reported record revenue and profit driven by soaring demand for its high bandwidth memory (HBM) solutions crucial for generative AI. The company’s HBM focus has made it a key player in the AI infrastructure market, with its 2026 memory product supply already fully committed. Revenue increased 39% and operating profit 62% year-over-year in Q3. SK Hynix aims to maintain its HBM leadership, supplying next-gen HBM4 chips. While competitors like Micron and Samsung are entering the HBM market, analysts project SK Hynix will preserve its dominant share throughout 2025.

TSMC reported a strong Q3, driven by the thriving AI chip market. Profit surged 39.1% year-over-year, with revenue exceeding expectations at NT$989.92 billion. Net income also surpassed estimates, reaching NT$452.3 billion. The company’s HPC division, fueled by AI and 5G infrastructure demands, was a key growth driver. Advanced chips using 7-nanometer or smaller technologies comprised 74% of wafer revenue, highlighting TSMC’s technological lead in miniaturization. TSMC’s performance underscores its critical role in the global semiconductor supply chain and the AI revolution.