Written by Ralph Sun
The semiconductor industry is defined by cycles of boom and bust, traditionally driven by the macroeconomic forces of consumer demand. However, the current artificial intelligence supercycle has introduced a new, unprecedented dynamic: a market where demand is effectively infinite, but the physical constraints of manufacturing are absolute.
On May 7, 2026, Arm Holdings (ARM) provided a stark illustration of this new reality. The British chip designer reported record revenue and earnings for its fiscal fourth quarter, easily beating Wall Street expectations. Yet, in a counterintuitive move, the stock plunged 10.1%, erasing more than $12 billion from its market valuation [1].
The culprit was not a failure of execution or a lack of customer interest. Instead, Arm’s stock collapsed because the company explicitly warned that it could not secure enough manufacturing capacity to fulfill the explosive demand for its newest AI-focused processors.
The AGI CPU Bottleneck
Arm’s business model is unique in the semiconductor ecosystem. The company does not manufacture physical chips; it designs the underlying architecture and licenses that intellectual property to companies like Apple, NVIDIA, and Qualcomm, collecting royalties on every chip shipped.
The immediate catalyst for the May 7 sell-off was commentary surrounding Arm’s new AGI CPU—a processor specifically engineered for agentic AI workloads. According to CEO Rene Haas, customer demand for this new architecture doubled from an initial target of $1 billion to over $2 billion in just six weeks [2].
However, the company maintained its flat revenue guidance for the chip. Why? Because Arm has only secured the foundry capacity to fulfill the first $1 billion tranche of orders; the manufacturing required for the next billion dollars of demand is not yet locked in [2].
Thesis on ARM: BUY
The 10% plunge in Arm shares represents a fundamental mispricing of risk by the market. The stock fell because of a supply constraint, not a demand problem. The underlying thesis for Arm remains intact: hyperscalers are rapidly adopting Arm-based server CPUs for their AI data centers due to their superior power efficiency. Management noted that the company now holds roughly 50% of the CPU compute market share among the largest cloud operators, including Amazon’s AWS and Alphabet’s Google Cloud [3].
While the valuation remains elevated at a price-to-earnings multiple of roughly 130 [3], the 10% dip offers a compelling entry point for long-term investors. The supply bottleneck is a temporary, solvable logistical issue; the structural shift toward Arm architecture in the data center is a permanent, multi-year tailwind.
The Super Micro Margin Squeeze
Arm’s supply constraints highlight the immense pressure placed on the entire AI hardware supply chain. Another prime example of this dynamic is Super Micro Computer (SMCI), a leading manufacturer of high-performance servers tailored for AI workloads.
Super Micro has been one of the most volatile stocks of 2026, experiencing massive single-day swings. On May 6, the stock surged 24.5% after reporting its fiscal third-quarter results [4]. However, the underlying financials reveal a deeply troubling trend: a severe trade-off between revenue growth and profitability.
The company missed its revenue expectations by over $2 billion, reporting $10.24 billion in sales [5]. Management attributed the miss to certain customers delaying deployments. Yet, the stock surged because adjusted gross margins spiked 370 basis points to 10.1% [5].
The celebration was short-lived. Super Micro’s guidance for the upcoming quarter projects sales rising to $11.8 billion, but gross margins falling back to 8.3% [5]. This inverse relationship between sales volume and profitability is a massive red flag in a high-growth industry.
Thesis on SMCI: SELL
Super Micro is caught in a margin trap, unable to scale its revenue without sacrificing profitability. Furthermore, the company is facing severe legal and reputational headwinds. The Department of Justice recently charged co-founder Liaw with allegedly conspiring to sell U.S.-made servers to China in violation of export controls [5]. While the company claims it is not the direct target of the investigation, the reputational damage is significant, especially following the 2024 resignation of its auditor, EY [5]. With a class-action lawsuit deadline approaching on May 26 [6], and major analysts like JPMorgan lowering their price targets [5], the risk-reward profile for Super Micro is heavily skewed to the downside.
The Broader Market Context
The supply constraints highlighted by Arm and the margin pressures faced by Super Micro are symptoms of a maturing AI hardware cycle. The initial phase of the buildout—characterized by indiscriminate buying of any company associated with AI—is over.
We are now entering the execution phase. The winners will be the companies that can navigate the treacherous waters of advanced semiconductor manufacturing, secure vital foundry capacity from TSMC and Samsung, and deliver complex systems at scale without destroying their profit margins.
Investors must become highly selective. A record earnings beat is no longer sufficient to guarantee stock appreciation; companies must prove they have the physical capacity to fulfill the demand they are generating. Arm Holdings possesses the structural advantages to overcome its current supply bottleneck, making its recent dip a buying opportunity. Conversely, Super Micro’s fundamental margin issues and legal liabilities make it a highly speculative, dangerous holding in an increasingly unforgiving market.
Disclaimer: This article is for informational purposes only and does not constitute investment advice. The opinions expressed are those of the author and do not reflect the views of Equities Orbis or its affiliates. Always conduct your own research and consult a licensed financial advisor before making investment decisions.
