Apple Inc. (Apple (AAPL)) is warning that rising memory and storage chip costs could force future product price increases, underscoring a broader shift in the AI investment landscape: the next wave of winners may be memory suppliers, not just GPU makers.
In an interview with The Wall Street Journal, Apple CEO Tim Cook said surging demand from AI data centers has tightened supply and driven up memory chip prices. Increasing production is being diverted to high-bandwidth memory (HBM), a specialized DRAM used in AI servers, reducing availability for consumer electronics.
The resulting squeeze is forcing companies like Apple to either absorb higher costs or pass them on to consumers.
AI's Next Bottleneck
For much of the AI boom, investors have focused on graphics processors and the companies supplying them. But as hyperscalers race to build AI infrastructure, memory has emerged as one of the industry's most critical bottlenecks.
HBM is essential for training and running advanced AI models because it allows processors to access massive amounts of data at high speeds. Demand has surged alongside AI spending, leading manufacturers to allocate more production toward data-center customers and away from traditional consumer markets.
Cook's comments suggest the supply-demand imbalance is now becoming visible beyond the semiconductor industry, with consumer technology companies beginning to feel the effects.
Are ETF Investors Looking in the Wrong Place?
The development could create opportunities for semiconductor ETFs such as the VanEck Semiconductor ETF (SMH), iShares Semiconductor ETF (SOXX), SPDR S&P Semiconductor ETF (XSD), and Strive U.S. Semiconductor ETF (SHOC), all of which hold companies benefiting from rising memory demand.
But the more intriguing ETF story may lie outside traditional AI funds.
The biggest beneficiaries of the HBM boom are South Korean memory giants Samsung Electronics and SK hynix, two companies that dominate the global memory market. Their growing importance in the AI supply chain could make country-specific funds such as the iShares MSCI South Korea ETF (EWY) and Franklin FTSE South Korea ETF (FLKR) indirect beneficiaries of continued AI infrastructure spending.
The development could also bolster interest in the Roundhill Memory ETF (DRAM), one of the newest funds designed to capitalize on growing demand for memory and storage technologies. Unlike broad semiconductor funds that are heavily influenced by chipmakers such as Nvidia, DRAM offers targeted exposure to companies at the center of the memory ecosystem, including Micron Technology Inc. (Micron (MU)), Samsung Electronics, SK hynix, Kioxia, and SanDisk Corp (SanDisk (SNDK)).
Apple's comments effectively validate the fund's core investment thesis that AI data centers are consuming increasing amounts of high-bandwidth memory, tightening supply across the broader market and giving memory manufacturers greater pricing power. As memory emerges as a critical bottleneck in the AI buildout, investors may increasingly look to specialized funds such as DRAM as a more direct way to play the trend.
Beyond Chips
Apple's warning also underscores how AI is creating opportunities across the broader infrastructure stack, from semiconductors and memory to power generation, networking equipment, and data centers.
While investors have poured billions into Nvidia-centric strategies, Cook's comments suggest the next phase of the AI trade may hinge on a less flashy but increasingly critical input: memory.
If AI data centers continue to absorb a growing share of global memory supply, chipmakers—and the ETFs that track them—could emerge as some of the biggest winners.
