Amid the AI infrastructure investment boom, memory chip stocks appear to benefit from surging computing power demand, yet two prominent investors and a tech analyst remain cautious. Their reasoning converges on the same core judgment: the high prices and high profits in memory are creating the forces that will ultimately destroy them.
Ark Invest founder Cathie Wood recently explained in a podcast why she does not hold memory stocks. She believes memory is the most commoditized and most cyclical segment of the semiconductor supply chain, and the current sharp price increases are not a normal state for the tech industry—they are essentially a negative signal.
Meanwhile, tech strategy analyst Ben Thompson used a sharper analogy, warning that memory makers are repeating Iran's mistake of blockading the Strait of Hormuz—effective in the short term, but in the long run it will push the entire industry to find an alternative route.
Their judgments directly challenge the current market sentiment that favors memory stocks. As AI inference demand expands rapidly, high-bandwidth memory (HBM) suppliers' share prices continue to attract buying, but the logic of these two investors suggests that investors should be wary of structural risks arising from technology route shifts and active demand-side avoidance.
Cathie Wood: High Prices Are a Warning, Not a Positive
Cathie Wood directly addressed questions about her decision to avoid memory stocks in a recent video. She admitted that years of investment experience may have made her more vigilant about cyclical industries.
In her view, memory is the most commoditized segment of the entire semiconductor supply chain, historically characterized by violent cyclical swings. The current trend of high-bandwidth memory prices tripling, quadrupling, or even rising tenfold is abnormal for the tech industry. "Most people think this is a huge positive, but it is actually a negative signal," she said.
She further supported this judgment from a cash flow perspective. She noted that a recent chart showed free cash flow for chip stocks moving in completely opposite directions to that of hyperscale cloud operators—the former benefiting while the latter comes under pressure. However, she emphasized that this state is temporary.
More critically, Cathie Wood pointed out that technological innovation is actively reducing reliance on high-bandwidth memory from the demand side. She cited Cerebras and Groq as examples—Ark's venture fund holds a stake in Groq—noting that the architecture designs of these two inference chip companies do not require high-bandwidth memory.
She compared this trend to Tesla's process of eliminating cobalt from batteries: once a supply chain component becomes too expensive or risky, engineers find ways to work around it. "In the inference space, we are seeing engineering technology replacing the need for high-bandwidth memory," she said.
Ben Thompson: Memory Makers Are Painting a Target on Their Own Backs
Tech strategy analyst Ben Thompson's concerns about the memory industry focus more on competitive dynamics. He used a geopolitical metaphor to describe the strategic dilemma memory makers currently face.
"I compare memory makers to Iran," Thompson said. His logic is that the deterrence of the Strait of Hormuz lies in the fact that it is always a card that can be played. Once it is actually used, it triggers the opponent's determination to bypass it entirely. "Now they have played that card, and it did work. But the UAE and Saudi Arabia will build oil pipelines and new ports so this never happens again."
He believes the current high-price strategy of memory makers is creating the same effect. On one hand, Apple is lobbying to bring in Chinese memory suppliers to break the existing supply structure; on the other hand, the primary optimization goal at the algorithm level has already become "how to reduce memory usage."
"I worry memory makers may have done the same thing," Thompson said. "No one will allow themselves to be put in this vulnerable position on memory again." His conclusion is that in the long run, memory makers may ultimately reap what they sow by creating such a massive "target."
Two Lines of Reasoning, One Conclusion
Although Cathie Wood and Ben Thompson use different analytical frameworks, their conclusions are highly consistent: the current strength of memory stocks is accelerating the process of their own replacement.
Cathie Wood approaches from the perspective of technology evolution, emphasizing that innovations in inference chip architecture will systematically reduce demand for high-bandwidth memory; Ben Thompson approaches from competitive dynamics and supply chain politics, pointing out that high prices and high concentration will drive buyers to seek alternatives—whether through algorithm optimization, bringing in new suppliers, or redesigning system architecture.
Their judgments collectively point to a risk worth noting for investors: within the AI infrastructure investment narrative, the benefit logic for memory stocks may be more fragile than the market expects, and the current boom may itself be the strongest force pushing the industry to accelerate its search for alternatives.
Comments