Wall Street fund manager: The most important company in the AI era is not Nvidia, but TSMC!

```

In the wave of investment in artificial intelligence, Nvidia has long been the focus of the market, but a fund manager overseeing nearly $10 billion in assets has offered a different answer: TSMC is currently the most indispensable core in the AI ecosystem.

According to a MarketWatch report on Tuesday, Jonathan Cofsky, co-portfolio manager of the Janus Henderson Global Technology and Innovation Fund, which manages $9.35 billion, said TSMC is the largest holding in his fund, surpassing Nvidia. "This may be the most important company driving AI development right now," he said. In his view, No matter which chip manufacturer ultimately prevails, all leading process chips are produced at TSMC. This irreplaceable status makes it the ultimate beneficiary of AI computing power expansion.

Cofsky’s fund has beaten the S&P 500 in both three- and ten-year annualized returns. He also points out that the current boom in AI infrastructure construction is essentially different from the internet bubble of the early 2000s—the physical constraints of tangible resources and electricity will effectively curb disorderly expansion of capital and provide support for this growth cycle.

TSMC: The "Only Answer" in the Computing Power Race

Cofsky attributes TSMC’s core value to a simple logic: investors only need to answer one question—"How much computing power will be built in the future?" And the answer is irrelevant to specific chip manufacturers, "because everyone is producing at TSMC."

TSMC is the world’s only wafer fab focused exclusively on contract manufacturing chips for other companies, possessing the most advanced process technology. Cofsky emphasizes that regardless of what form AI computing takes, TSMC is an unavoidable production node, giving it a unique structural advantage in the entire AI industry chain.

Betting on Arm: A New Beneficiary in Agentic AI Workflows

Besides TSMC, Cofsky revealed that his fund increased its position in chip architecture design company Arm Holdings earlier this year. The reason is that agentic AI workflows rely far more heavily on CPUs (central processing units) than previous AI iterations, and Arm boasts "the widest architectural penetration."

He uses an illustrative analogy to explain the division of labor between GPUs and CPUs: GPUs are like the "army of workers" executing tasks, while CPUs are the "managers" responsible for overall scheduling; agentic AI involves a lot of "scheduling combinations," which is Arm’s strong suit.

Cofsky also notes that Arm is transitioning from a pure IP licensing model to manufacturing its own chips, with Meta as its first client. "By the end of this decade, Arm’s success as a chip maker will be a key variable to watch."

Software Stocks: Only 10% to 20% of Current Players Will Survive the Cycle

Discussing the recent rebound in software stocks, Cofsky remains cautious. He believes the software industry faces disruptive risks from AI, since AI fundamentally automates labor and replaces manpower.

He cites historical patterns, noting that in every disruptive long-term trend—for instance, retail shifting from offline to online platforms like Amazon—it usually takes four to six years for the market to distinguish winners and losers, and ultimately only 10% to 20% of original leading companies succeed in transforming.

Based on this, the fund focuses on companies that are "extremely difficult to replicate and can overlay AI as incremental revenue," including Datadog, Snowflake, and theoretical and design software firms with unique data assets such as Cadence and Guidewire. "Snowflake and Datadog have already seen re-acceleration of revenue thanks to AI, which is the most positive signal we can see so far," he said. "But we need to observe whether this momentum can continue."

Contrarian Layout: DoorDash’s Underestimated Logic

Cofsky also named one currently less popular stock—food delivery platform DoorDash. He believes that market concerns are currently focused on DoorDash’s ongoing investment cycle and macroeconomic pressure on consumers, but these factors are overly priced in.

In his view, DoorDash is a founder-driven company actively expanding into international fresh grocery and other new businesses, and is expected to benefit from both generative AI and physical AI, making it valuable for medium- to long-term positioning.

Risk Warning and DisclaimerThe market carries risks, and investment should be approached with caution. This article does not constitute personal investment advice, nor does it take into account the specific investment goals, financial situation, or needs of any individual user. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular circumstances. All investment decisions made based on this article are the responsibility of the user. ```