Seoul’s Stance: Memory Is The Key Limiting Factor For AI Progress

South Korea’s SK hynix has publicly warned that memory supply cannot keep pace with surging AI demand, with no significant new capacity expected in 2026. This development highlights a looming global memory shortage that could hinder AI progress and intensify geopolitical tensions, as governments and corporations vie for limited high-bandwidth memory resources.

During a press briefing at the Korea Chamber of Commerce and Industry’s Jeju Forum, SK hynix chairman Chey Tae-won stated that customers are requesting 60 to 100 percent more AI memory in 2027 than they are currently receiving. Chey estimated that AI now accounts for over half of total semiconductor consumption, with demand growth projected at a minimum of 50-60 percent.

He emphasized that No company has meaningful new capacity coming online next year, highlighting a significant supply-demand imbalance. This shortage is most acute in high-bandwidth memory (HBM), crucial for AI accelerators, which is leading to what Chey described as near-chaotic lobbying from corporate and government actors. Governments are increasingly viewing memory access as an issue of economic security, intervening to protect domestic industries.

Chey warned that more concentrated control over memory capacity could lead to geopolitical conflicts, with governments pressuring each other as demand outstrips supply. SK hynix currently holds approximately 58 percent of the global HBM revenue, underscoring the industry’s oligopolistic structure.

In response, SK hynix announced plans to accelerate capacity expansion, including moving forward the opening of its Yongin mega-cluster to February 2027 and committing over $14 billion in new investments. However, none of this capacity will be available in 2026, meaning a supply gap is already locked in for the immediate future.

Implications of Memory Shortages for AI Development and Geopolitics

This warning from SK hynix underscores a critical bottleneck in AI advancement, as the shortage of high-bandwidth memory could slow the deployment of larger, more capable AI models. The concentration of memory capacity in a few firms and regions raises concerns about geopolitical vulnerabilities, especially as governments begin to treat memory access as a matter of national security. For AI developers and policymakers, the shortage could lead to increased costs, delays, and intensified competition for limited resources, shaping the future landscape of AI innovation and international relations.

Rising Demand and Limited Capacity in AI Memory Supply

The current memory shortage is rooted in a rapid increase in AI compute demands, with AI now representing over 50% of semiconductor consumption. SK hynix, Micron, and Samsung dominate the high-bandwidth memory market, with SK hynix holding a majority share. Despite recent investments, capacity expansions are not expected to meet the surging demand in 2026, creating an ongoing imbalance.

Chey Tae-won’s remarks follow a pattern of warnings about the risks of high memory prices, which he describes as abnormal and potentially harmful to the industry. The industry’s physics dictate that new capacity will only arrive gradually, leaving a lock-in of shortages in the immediate term. Additionally, the geopolitical landscape is shifting, with governments increasingly involved in memory allocation debates, viewing access as a matter of economic security.

“No company has meaningful new capacity coming online next year.”

— Chey Tae-won, SK hynix Chairman

Unconfirmed Aspects of Future Capacity and Market Response

It remains unclear how quickly SK hynix and other suppliers will accelerate capacity expansion and whether new capacity will be sufficient to meet the projected demand growth. The specific timeline for resolving the shortage and the potential for government intervention or new geopolitical tensions are still developing issues.

Next Steps in Capacity Expansion and Policy Responses

SK hynix plans to accelerate its capacity expansion, with new plants expected to come online in 2027. Industry analysts will monitor whether these investments can bridge the demand gap. Additionally, governments may increase intervention, potentially restricting exports or providing subsidies, which could reshape the global memory market and AI development landscape.

Key Questions

How will the memory shortage affect AI development?

The shortage could slow the deployment of larger AI models, increase costs, and create delays in AI innovation, especially at the frontier scale requiring high-bandwidth memory.

Why is memory capacity concentrated among few companies?

The high costs and physics of manufacturing high-bandwidth memory have led to a market dominated by three firms, creating an oligopoly that intensifies supply risks.

What geopolitical risks are associated with memory shortages?

Governments are increasingly viewing memory access as a matter of national security, which could lead to export restrictions, tariffs, or other interventions affecting global supply chains.

Will capacity expansion solve the shortage?

While SK hynix and others are investing heavily, capacity will not be sufficient in 2026, meaning shortages are expected to persist into 2027 and beyond.

How does this impact consumers and businesses?

Memory shortages can lead to higher prices for consumer electronics and enterprise hardware, as well as increased costs and delays for AI applications and infrastructure.

Source: ThorstenMeyerAI.com