Jensen Huang's Dual Endorsements Reshape AI Chip Investing
Nvidia CEO Jensen Huang has set the semiconductor world abuzz with a series of high-profile comments this week, offering investors a rare glimpse into the expanding artificial intelligence (AI) landscape. During a recent visit to Seoul, South Korea, Huang openly praised Qualcomm (QCOM), calling it a smart buy for those looking to profit from AI in mobile devices and edge computing.
"I don't think we're incredibly good at mobile devices, and I don't think it's necessary," Huang said, directing investors toward Qualcomm instead. The endorsement, reported by The Motley Fool, comes as Qualcomm shares have already gained 4.28% in recent trading, with a market cap of $214 billion. Huang acknowledged Qualcomm's success in on-device AI, where chips must deliver inference while drawing minimal power to preserve battery life.
Marvell Technology: The Next Trillion-Dollar Bet?
Just days earlier, Huang caused a stir at Computex 2026 in Taipei when he suggested that Marvell Technology (MRVL) could become the next $1 trillion company, according to a Yahoo Finance report. Huang emphasized that Marvell's networking and connectivity chips are essential for handling AI workloads as data centers scale up, calling high-speed connectivity a critical bottleneck. His endorsement followed a strategic partnership in March where Nvidia invested $2 billion in Marvell to integrate its technology into the Nvidia NVLink Fusion ecosystem.
Marvell, with a current market cap of about $245 billion, has seen its stock surge on the back of Huang's comments. The company's revenue rose 41% to $8.2 billion last financial year, with analysts forecasting another 40% growth this year. Marvell's custom AI chips and optical interconnects that enable massive GPU clusters to function as single unified entities have positioned it as a key player in the AI infrastructure boom.
Why It Matters: AI's Expanding Footprint Beyond Data Centers
Huang's dual endorsements highlight a strategic shift in the AI landscape. While Nvidia's high-performance GPUs dominate data center training workloads, the Nvidia CEO recognizes that not every pocket of the AI realm requires such powerful hardware. Qualcomm's Snapdragon chip platform is marketed toward on-device processing across smartphones, laptops, automotive systems, and IoT devices, prioritizing low latency and energy efficiency.
Qualcomm has also aggressively expanded into AI inference for data centers with its AI200 and AI250 accelerators, optimized for lower power consumption and memory management. This architecture directly targets workloads where Nvidia's ecosystem may be too costly. Huang's acknowledgment that Nvidia lacks competitive advantages in edge devices underscores the growing specialization within the semiconductor industry.
TSMC: The Unseen Beneficiary
In a separate development reported by Barchart, Huang called AI "insanely profitable" for TSMC (TSM), the Taiwan-based chip manufacturer that produces Nvidia's advanced GPUs. TSMC has been a critical partner in Nvidia's meteoric rise, and Huang's comments reinforce the symbiotic relationship between the two companies. As AI demand surges across data centers, edge devices, and automotive applications, TSMC's advanced manufacturing processes become even more valuable.
Perspective: The Broader Implications of Huang's Vision
Huang's recent comments paint a picture of an AI industry that is diversifying rapidly. While Nvidia remains the dominant force in accelerated computing for data centers, the future of AI will be built on multiple specialized chips optimized for different tasks. Qualcomm's success in mobile AI and Marvell's leadership in networking and custom ASICs demonstrate that the AI revolution extends far beyond GPU clusters.
A Philosophical Note: Time and Dedication
Beyond the business implications, Huang's recent interviews offer a glimpse into his personal philosophy. Speaking with The Times of India, the Nvidia CEO revealed that he doesn't wear a watch, explaining that "now is the most important time. Just dedicate yourself to now." He shared a story about a gardener in Kyoto who had spent 30 years perfecting a moss garden with bamboo tweezers, telling Huang, "I have plenty of time."
This philosophy of patience and dedication has clearly served Huang well. At 63, he leads one of the world's most valuable tech empires, with a workforce largely composed of millionaires. His journey from clearing tables at Denny's to becoming the fourth most successful living immigrant in America (according to Forbes' recent list for the nation's 250th anniversary) is a testament to the power of focus and long-term vision.
What This Changes for Investors
Huang's endorsements of Qualcomm and Marvell signal that the AI investment opportunity is broadening. While Nvidia remains the bellwether, investors looking for exposure to AI should consider companies that complement Nvidia's strengths rather than compete directly. Qualcomm's edge computing capabilities and Marvell's networking solutions represent distinct niches that are likely to grow alongside Nvidia's core business.
For those wondering about the broader economic context, the AI boom continues to reshape industries, even as other sectors face disruptions. The current global landscape, marked by events such as the airline shutdown that strands thousands, underscores the uneven nature of technological progress. Yet for investors willing to look past near-term volatility, Huang's latest comments offer a clear roadmap: the AI future is not monolithic, and specialized chipmakers like Qualcomm and Marvell are poised to play critical roles.
Comments