Jensen Huang didn’t just steal the spotlight at Computex 2025—he became it.
Amid the leather-jacketed frenzy dubbed “Jensanity”, the 62-year-old Nvidia co-founder surfed a tidal wave of Taiwanese adoration, made billion-dollar announcements, and left behind a not-so-subtle message: Nvidia is betting everything on AI—and plans to stay king no matter what.
But even kings face headwinds. While Nvidia’s valuation soars past $2.5 trillion and Huang’s personal net worth climbs beyond $100 billion, the company is confronting a more complicated global stage: potential AI infrastructure spending cuts, geopolitical chokepoints, and a saturated customer base.
Still, if anyone can turn tectonic shifts into golden opportunities, it’s the man who turned GPUs into the engine of the AI age.
👑 The Gospel of Jensen
To say Huang is bullish on AI is like saying TSMC dabbles in silicon. The Nvidia CEO sees artificial intelligence as the defining industry of our era, comparing its transformative potential to the invention of electricity or the Internet. But instead of offering an abstract vision, Huang brings a blueprint: build the infrastructure, power the data centers, and dominate the entire stack.
He’s not just a CEO—he’s Nvidia’s founding prophet. His personal wealth is tied tightly to the company (he still owns 3.5%), and even his kids work there. If his AI empire keeps growing, he could realistically become one of the wealthiest people in history.
His belief? AI isn’t optional. It’s inevitable.
“AI infrastructure is going to be part of society,” Huang told Reuters. “That’s why I’m traveling around the world.”
It’s not just sales talk—it’s strategy.
🌀 Trouble in the AI Cloud
Behind the curtain of standing ovations and guava gifts lies a tougher reality: Nvidia’s biggest cloud customers are hitting pause.
Microsoft and Google have both signaled cutbacks in AI infrastructure spending. The golden days of hyperscaler megadeals may be ending—not every country can throw $10–50 billion at sovereign AI data centers like Saudi Arabia.
As Seaport Research’s Jay Goldberg bluntly put it:
“They’re sort of running out of obvious deals.”
And then there’s China. U.S. export restrictions have forced Nvidia to downgrade or withdraw chips for the Chinese market, handing opportunities to domestic players and threatening one of its largest international customer bases.
In short, Nvidia’s growth model is being stress-tested.
🔌 NVLink Fusion: The New Playbook
To stay ahead, Nvidia is shifting from sovereign AI plays to something more modular—and arguably more brilliant.
Enter NVLink Fusion: a new interconnect tech that lets third-party companies plug custom chips directly into Nvidia’s AI infrastructure.
Think of it as Intel Inside’s strategy for the AI era—except this time, Nvidia owns the rack, the silicon highway, and the market pull.
Instead of building their own servers from scratch, partners can innovate on smaller components and still tap into Nvidia’s massive software and hardware ecosystem.
“It creates a platform effect,” said Marvell VP Nick Kucharewski. “It lowers the barrier to entry.”
In Huang’s playbook, every company becomes an Nvidia company—even if they don’t know it yet.
🏢 Enterprise AI: The Final Frontier?
This week, Nvidia also debuted its new line of enterprise AI supercomputers, capable of handling everything from virtual machines to generative workloads. The goal: tap into a multi-billion-dollar market beyond Big Tech and Big Government.
But cracking the enterprise market is notoriously slow. Deals are small, diverse, and require trust-building over time—nothing like headline-grabbing sovereign AI city-states.
As analyst Goldberg noted:
“We’re bumping up against the limits of expanding the customer base.”
Still, Huang is undeterred. He sees scale everywhere, and he’s betting that the infrastructure race will be global, granular, and perpetual.
🧬 Taiwan: The Beating Heart of the AI Supply Chain
Nvidia’s deep ties with Taiwan are no accident—they’re essential. TSMC manufactures Nvidia’s most advanced chips. Hundreds of Taiwanese firms supply their AI servers with boards to thermal modules.
“The purpose of Computex was to bring together the ecosystem and the supply chain,” said analyst Ian Cutress.
This week proved that Taiwan isn’t just a foundry. It’s the nervous system of the AI revolution—and Huang is its spiritual and economic emissary.
Local affection for Huang runs deep. Born in Tainan, Taiwan’s ancient capital, before emigrating to the U.S. at age 9, he’s widely viewed as a native son made good. This week, he dined with Foxconn’s Young Liu, shared fruit with MediaTek’s Rick Tsai, and turned startups like Solomon Technology into overnight stock darlings just by name-dropping them.
Solomon’s chairman, Johnny Chen, said:
“We weren’t very well-known before, but with Nvidia’s support, many more people know us now.”
Indeed—Jensen speaks, markets move.
🧠 Nvidia’s Empire Is Shifting Gears, Not Slowing Down
If this week proved anything, it’s that Nvidia has graduated from chipmaker to civilization builder. And while some clouds gather on the AI horizon—regulatory, financial, and geopolitical—Huang has built an empire with momentum, mindshare, and mission.
The pivot to NVLink Fusion is bold and timely. It invites innovation while preserving Nvidia’s control of the rails. The enterprise push shows hunger, even if it’s uphill. And the Taiwan partnership? That’s no accident—it’s the secret sauce behind every AI server sold.
Yes, growth may moderate. But don’t mistake that for decline. Nvidia isn’t reacting to the AI slowdown. It’s redefining the next phase—from hyperscaler deals to platform lock-in to enterprise ubiquity.
If AI is the new electricity, Nvidia doesn’t just want to sell you the lightbulbs. It wants to own the grid.
