AI infrastructure is consuming so much computer memory that automakers, medical device manufacturers, and telecommunications companies are facing critical shortages and rising costs. A coalition of nine major US trade associations sent an urgent letter to Commerce Secretary Howard Lutnick and Treasury Secretary Scott Bessent on June 3, warning that the explosive growth of AI data centers is creating a structural shift in the global memory market that threatens industries far removed from artificial intelligence.

Why Are AI Data Centers Hoarding Memory Chips?

The problem centers on a specialized type of memory called high-bandwidth memory, or HBM. This is the ultra-fast memory that powers AI accelerators from companies like Nvidia and AMD. As hyperscalers, the massive cloud computing companies that run AI services, race to deploy larger AI clusters, they are demanding an unprecedented share of global memory production.

Samsung and SK Hynix, which together control over 95% of global DRAM (dynamic random-access memory) production, have been diverting their manufacturing capacity toward high-margin HBM for AI accelerators. This shift is starving the commodity DRAM and NAND markets, which supply memory for consumer electronics, broadband infrastructure, automobiles, and medical devices. Both companies warned in April that significant shortages will continue through at least 2027.

The impact is already visible. IDC, a market research firm, has revised its 2026 personal computer market forecast downward by up to 9% as a direct consequence of memory scarcity and rising prices.

What Industries Are Being Hit Hardest?

The coalition's letter identifies several sectors facing disruption from the memory shortage:

• Automotive: Disruptions to automobile production as memory supplies tighten and costs rise.
• Medical Devices: Delays affecting the manufacturing of medical equipment that relies on memory chips.
• Telecommunications: Increased costs for broadband and telecommunications infrastructure expansion.
• Consumer Electronics: Higher prices for personal computers, smartphones, and other devices that depend on standard memory chips.
• Federal Contractors: Delays affecting government contractors attempting to fulfill procurement obligations.

The coalition acknowledged AI's importance but argued it should not come at the expense of the rest of the economy. "While recent developments in AI offer the promise of generational technological advances and are important for US tech leadership, we must also ensure other key industries are not negatively impacted by this disruption in the marketplace," the organizations stated.

How Can Policymakers Address the Memory Shortage?

The coalition has proposed several concrete steps for federal intervention. These recommendations aim to rebalance the memory market and protect non-AI industries from sustained supply constraints:

• Accelerate Domestic Expansion: Speed up the expansion of memory manufacturing capacity in the United States and allied nations to increase overall supply.
• Strengthen Supply Chain Resilience: Use trade agreements to build more resilient semiconductor supply chains and reduce dependence on single suppliers.
• Ensure Equitable Allocation: Work directly with memory suppliers and major chip buyers to ensure adequate memory supply for non-AI industries.
• Leverage Existing Programs: Use CHIPS Act programs, which provide federal funding for semiconductor manufacturing, to support memory capacity growth.
• Reduce Regulatory Barriers: Remove regulatory obstacles that may slow capacity expansion and production timelines.

The coalition urged the administration to "work with memory chipmakers and chip buyers to assess steps that can be taken to address this imbalance in the memory market and protect against harm to consumers, workers, and businesses of all sizes".

Is This a Temporary Problem or a Permanent Shift?

Industry analysts have repeatedly warned that AI demand is reshaping the economics of the memory market in a fundamental way. While memory shortages have historically been cyclical, driven by temporary supply-demand imbalances, the coalition argues that AI infrastructure spending is creating a structural shift large enough to affect industries far removed from data centers.

The letter marks the first coordinated, multi-industry push for federal intervention on the memory shortage. Whether the Trump administration will respond, and how quickly, remains to be seen. The stakes are high: the shortage threatens not just consumer prices but also the ability of critical industries like automotive and medical device manufacturing to maintain production schedules.

Meanwhile, the competition for memory continues to intensify. Qualcomm has recently announced plans to expand into AI data center chips with its new Dragonfly brand, signaling that demand for AI-specific hardware will only grow in the coming years. This suggests the memory crunch could persist unless manufacturing capacity expands significantly.