Chip Stocks on Track for Best Quarter Ever

2026-06-30 · Source: Bloomberg Tech · Field: Business & Management — Corporate Strategy & Leadership, Entrepreneurship & Start-ups, International Business & Trade · Depth: Fundamental Awareness, extended

Summary

The Philadelphia Semiconductor Index (SOX) is experiencing its best quarter ever, up 86% in three months and nearly 100% year-to-date by June 30, 2026, driven by robust AI infrastructure spending. This rally is significantly fueled by demand for memory and storage components from companies like Micron, which reported an 85% gross margin. While volatility exists, demand is projected to remain elevated into 2027-2028. Beyond chips, the AI build-out requires substantial investment in networking, cooling, power, and electricity. Separately, Taiwanese authorities are investigating Super Micro for alleged Nvidia chip smuggling to China, violating US export controls. US lobbying firms are also severing ties with Chinese tech giants due to new restrictions. Startup Etched, focused on low-latency AI inference systems, raised \$800 million. The Supreme Court rejected birthright citizenship restrictions, yet US entry-level visa sponsorship has plummeted to 2.3%. Comcast plans to split NBCUniversal and Sky.

Key takeaway

For investors evaluating the AI market's long-term sustainability, recognize that the current chip rally, particularly in memory and storage, is robust but also highly volatile. Diversify your portfolio beyond core GPU manufacturers to include companies providing essential AI infrastructure like memory, networking, cooling, and electricity. Be aware of geopolitical risks, such as export controls and lobbying shifts, which could impact supply chains and market access for major tech players.

Key insights

AI infrastructure demand is driving historic chip market growth, shifting focus to memory, power, and specialized inference systems.

Principles

• AI infrastructure spending fuels broad tech sector growth.
• Memory and storage are critical bottlenecks in AI data centers.
• Specialized inference hardware can optimize cost-per-token.

Method

Etched's rack-scale inference system uses low voltage inference and cluster scale memory to reduce power consumption and increase user capacity on HBM.

In practice

• Consider memory and power providers for AI infrastructure investments.
• Explore low-voltage inference systems for cost-efficient AI deployment.

Topics

• Semiconductor Market
• AI Infrastructure
• Memory Chips
• US-China Tech Rivalry
• AI Inference Hardware
• Immigration Policy

Best for: CTO, Entrepreneur, Investor, Executive, Consultant

Related on AIssential

• AMD Soars on Blockbuster AI-Fueled Forecast — AI demand is driving significant growth and strategic shifts across semiconductor, cloud, and consumer tech sectors.
• Micron, Qualcomm Show Strong AI Demand — Surging AI demand is reshaping semiconductor markets, driving memory price hikes and strategic tech diversification.
• Nvidia’s Share of AI Inference Chip Market Appears to Be Rising — Nvidia's share in the AI inference chip market is reportedly growing.
• Why the Memory Crunch Is Almost Impossible to Solve — The global memory-chip shortage is intractable due to concentrated production, long factory lead times, and surging AI demand, compounded by geopolitical constraints.
• ChinAI #348: China's Compute Year in Review - frenzy, growing pains, and key milestones — Early hype for "all-in-one machines" and compute leasing in China's AI sector quickly faded due to practical limitations and fraudulent practices.
• Sector Snapshot: Semiconductor Startup Funding Still Running Hot — Semiconductor startup funding remains robust, driven by AI infrastructure demand, with significant rounds continuing despite public market fluctuations.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Bloomberg Tech.