However, the official cautioned that a closer look at the generative AI infrastructure theme reveals that markets are still in the early stages of understanding its potential, highlighting the fact that less than 2% of 12 million servers sold in 2023 were AI-enabled.

“A look at the global server environment shows that this theme is very nascent,” said Bradley Amoils, managing director and portfolio manager of US-based Axiom Investors, which manages the Pengana Axiom International Fund for Australian retail investors.

The term AI servers refers to servers specifically designed to handle the demands of AI workloads and they play an important role in the AI ​​technology landscape, supporting AI workloads throughout along the pipeline – from data preparation and training/fine-tuning to deployment and ongoing management. , according to Intel.

Amoils noted that even with Nvidia’s planned sale of 450,000 AI-enabled servers in 2024, these would still constitute only a small fraction of the global supply.

“If we look further, some suggest that AI-based servers could account for 5.3% of servers sold in 2026, which would still be well below demand, and suggest that we are only at the beginning,” he added.

Despite this, Amoils believes that Nvidia remains well positioned to generate strong revenue from AI generation in the coming quarters.

“It’s hard to say how high Nvidia can go, and we don’t like to predict these things, but if we look at forecasted revenues for the next few quarters, there’s no sign of this trend slowing down,” he said. he declared.

The Axiom executive also reminded investors that producing a chip takes about three to four months, noting that “if a chip goes into the foundry this quarter, it generates revenue the next quarter.”

He believes this makes Nvidia-related stocks, such as foundry partner TSMC and Vertiv, attractive to investors.

“These companies provide interesting data and show that the AI ​​theme extends from large-cap positions all the way to small-cap positions,” he added.

Two rounds of AI

Other fund managers share the view that although the current AI trend may be “overblown”, the long-term transformative potential of AI and investment opportunities in AI are valid.

Capital Group portfolio manager Mark Casey noted that markets tend to overestimate megatrends in the short term while underestimating them in the long term.

“What is remarkable about AI is its broad potential utility. Because it can take on a multitude of human tasks, I consider the AI ​​market to be of unknowable scale,” he said.

Another portfolio manager at Capital Group, Cheryl Frank, expects overcapacity and surpluses in the near term as companies still experiment with AI and determine how to use it to gain a competitive advantage.

However, Frank warned that the lack of adequate electrical capacity, basic materials or capital goods could slow AI’s expansion.

“I think there will be two AI cycles,” Frank said, noting that currently the market is in the middle of an advertising-driven consumer AI cycle, which will likely see a pullback.

“But the enterprise AI cycle will be much longer and slower,” he noted.

Discussing trends and investment opportunities in AI, Capital Group highlighted the “four-layer technology stack,” which enables AI workloads and supply chains required for infrastructure of AI, as one of the key opportunities where an AI spending spree could pay off. investors.

Depending on the company, some companies will capitalize on each layer of the stack, while others will release products across two or three layers.

“The question is which companies will perform best and which will stumble. That’s what I’m focused on,” Casey added.

The AI ​​stack includes semiconductors, cloud infrastructure, large language models such as Chat GPT, and end-user applications.

Meanwhile, chipmakers like NVIDIA and ASML operate on a single layer while tech giants like Alphabet, Microsoft and Amazon seek to dominate multiple layers.