Financial Directions March 12, 2025 3

How to Tackle Slowed Data Center Chip Demand?

Advertisements

The semiconductor industry is navigating a complex landscape as global investments in data centers show signs of potentially slowing downAmid these uncertainties, the demand for artificial intelligence (AI) smartphones could emerge as a vital growth driver, as suggested by Doug Lefever, the CEO of Advantest, a leading global semiconductor testing equipment manufacturerOn December 26, he highlighted that the surge in AI smartphone demand may help bolster the semiconductor industry and avert a potentially severe downturn.

Lefever elaborated on the contrasting yet significant investment patterns seen in major technology companies such as Meta, Google, and MicrosoftWhile their foray into AI remains robust, their investments in data centers appear to be waning, a trend that, if it persists, could ripple through the semiconductor supply chainThe concentration of control within the massive data center market means that even a slight slowdown could have major implications for semiconductor manufacturers

While Lefever stops short of describing this situation as a “bubble,” he warns of the potential for intensified cyclical fluctuations, underscoring the unyielding challenges the industry may face as it braces for the next economic cycle.

In an encouraging note, despite the prevailing uncertainties around data center investment, Lefever’s outlook on AI smartphone demand remains optimisticHe noted that, while current demand might be lackluster, there exists a possibility for explosive growth in the future“Everyone is holding their breath, waiting for the killer application for AI smartphones to arrive,” he explainedIf such applications materialize and consumers begin to replace their phones en masse, the market could witness unprecedented growth.

On the global stage, competition in the AI smartphone arena is heating upFor instance, Samsung recently introduced three new AI-enabled smartphones—S24, S24 Plus, and S24 Ultra—boasting an array of AI features such as real-time call translations, voice recording transcriptions, video searches, and advanced photo editing capabilities

With prices ranging from $800 to $1,300 in the United States, Samsung's strategic entry into AI smartphones aims to invigorate the mobile market and differentiate itself from rivals like Apple.

In addition to Samsung’s initiatives, Google is working on integrating local AI technology within its Pixel series, while domestic manufacturers such as Honor, OPPO, and Xiaomi are also making strides in launching AI-capable smartphonesAccording to market research firm Canalys, AI functionalities are expected to become a central selling point for smartphones moving forward, projecting that 5% of smartphone shipments this year will incorporate AI features, with the number potentially soaring to 635 million units or 45% of the total smartphone market share by 2027.

Despite the buzz around AI smartphones, some analysts remain cautious about whether the latest AI functionalities will trigger widespread upgrades among consumers

Analysts have pointed out that many existing smartphones already come equipped with certain AI features, indicating that Samsung's newly introduced capabilities may not be sufficient to spark a significant replacement wave.

Park Kang-ho, an analyst at Daishin Securities, echoed these sentiments by suggesting that the current AI offerings may not be compelling enough to motivate consumers to switch their devices en masseHowever, he added a silver lining, proposing that if AI smartphones can achieve deeper interconnectivity with other devices—such as vehicles, appliances, and computers—it could potentially catalyze broader IT device demand and, in turn, foster overall market growth.

Amidst these unfolding industry dynamics, Advantest stands to gain significantly from the rapid uptick in semiconductor demandHeadquartered in Tokyo, Advantest is recognized as a forefront manufacturer of semiconductor testing equipment, playing a pivotal role in delivering high-end graphics processing unit (GPU) test devices for clients like Nvidia

alefox

As semiconductor technology continues to advance, the appetite for sophisticated testing equipment has skyrocketedLefever noted that as chips grow in complexity and cost, the demand for testing services is experiencing explosive growth.

Recent performance indicators underscore Advantest’s substantial progress, with the company’s stock price surging over 80% in the past year and witnessing a staggering 500% increase over the past five yearsSuch remarkable growth reinforces the market's recognition of Advantest's dominance in semiconductor testing.

Even though Japan has lost its preeminence in global chip production since the 1980s, companies like Advantest remain crucial players in the global semiconductor supply chainOperating within niche markets, these organizations contribute to essential components and services that support the seamless functioning of the global semiconductor ecosystem.

Lefever emphasized that modern chip manufacturing necessitates extensive testing at various stages

Each chip typically undergoes between 10 to 20 tests throughout the process—from wafer slicing to final assembly—significantly up from the lower single digits seen five years agoThis trend highlights the increasing importance of quality control in chip production, driving an insatiable demand for advanced testing equipment.

Additionally, extending testing times serves as another key factor contributing to Advantest's successFor example, Nvidia's latest Blackwell products require testing durations three to four times longer than their predecessors, illustrating a growing complexity in chip design and production that demands meticulous testing.

Buoyed by surging semiconductor demand, Advantest has substantially revised its performance outlook for the 2024 fiscal year, now anticipating a 16% increase in net profit to reach 122 billion Japanese yen (approximately $792 million). Lefever expressed confidence, stating, “We are optimistic about future demand, particularly in relation to AI

Post Comment

Your email address will not be published. Required fields are marked *+