Signal: Memory Is The Quieter Chokepoint — And Seoul Just Said So Out Loud
AIThis post was created with the assistance of artificial intelligence (AI).

TL;DR

AUDIBLE

Listen free for 30 days with Audible

Thousands of audiobooks and originals — cancel anytime.

Start your free trial

As an affiliate, we earn on qualifying purchases.

South Korea’s SK hynix CEO publicly stated that the memory industry faces a significant supply shortfall, with no new capacity expected in 2026. This acknowledgment underscores a looming bottleneck driven by surging AI memory needs and geopolitical tensions.

South Korea’s SK hynix CEO Chey Tae-won has publicly warned that there will be no meaningful new capacity coming online in 2026 to meet the rapidly increasing demand for AI memory, which he estimates will grow by 50–60% in 2027. This marks a rare acknowledgment from industry leadership of a looming memory supply shortfall amid escalating geopolitical tensions and market pressures.

During a press briefing at the Korea Chamber of Commerce and Industry’s Jeju Forum, Chey Tae-won highlighted that customers are demanding 60–100% more AI memory in 2027 than this year. With AI now accounting for over half of total semiconductor consumption, the industry faces an unprecedented demand-supply imbalance. Chey emphasized that No company has meaningful new capacity coming online next year, signaling a looming capacity gap.

He also described the current high memory prices as unusual, warning they could lead to chip inflation and attract new entrants into the market, including Elon Musk’s semiconductor initiatives. Chey’s remarks suggest that the industry’s current margins are a strategic concern, as monopolistic control over high-bandwidth memory (HBM) remains concentrated among three firms—SK hynix, Micron, and Samsung—who hold the majority of global revenue. SK hynix announced plans to accelerate capacity expansion, including a new HBM-focused plant slated for February 2027, but this will not impact supply in 2026, creating a capacity gap.

At a glance
breakingWhen: announced July 2026
The developmentSeoul’s SK hynix CEO publicly confirmed that the global memory supply will not meet the rising AI demand in 2026, highlighting a critical bottleneck and geopolitical implications.

Implications of Memory Shortage for AI and Geopolitics

This public acknowledgment indicates that the industry anticipates a potential mismatch between demand and supply for high-bandwidth memory, which could affect AI infrastructure development. The shortage may influence costs for device manufacturers and consumers and could have geopolitical implications, given the concentration of memory production among a limited number of firms and regions. Chey’s comments highlight the importance of securing memory supply chains within the broader context of AI advancement and economic stability.

Amazon

High bandwidth memory (HBM) modules

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Memory Industry Dynamics and Geopolitical Tensions

The global memory market is primarily controlled by three companies—SK hynix, Samsung, and Micron—who together account for over 90% of high-bandwidth memory revenue, with SK hynix holding approximately 58% as of Q1 2026. Despite increasing demand, no significant new capacity is expected to come online until 2027, resulting in a capacity shortfall that has been recognized within the industry for some time. This situation is further influenced by geopolitical considerations, as governments are increasingly viewing memory access as a matter of economic security and may intervene to protect domestic industries.

Chey Tae-won’s remarks represent a notable departure from typical industry silence regarding capacity constraints, emphasizing that current high prices and demand growth are unsustainable. His comments about potential government interventions and lobbying efforts reflect the complex geopolitical environment surrounding supply security.

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

— Chey Tae-won, SK hynix Chairman

Amazon

AI memory modules for servers

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Unconfirmed Aspects of Capacity Expansion and Market Response

Although SK hynix has announced plans to accelerate capacity, including a new plant scheduled for February 2027, it remains uncertain whether these measures will fully meet future demand or if other industry players will undertake additional capacity expansions. The timeline for supply adjustments and the potential influence of government policies are still unclear, and the overall impact of these developments has yet to be determined.

Amazon

DDR5 RAM for high-performance computing

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Monitoring Capacity Developments and Market Responses

Industry analysts and market participants will observe SK hynix’s capacity expansion efforts and any additional investments by other memory manufacturers. Additionally, government actions aimed at securing memory supply chains or restricting exports could influence market dynamics. The industry’s response to Chey’s comments will be important in shaping the memory market outlook through 2026 and beyond.

Amazon

Memory upgrade for gaming PCs

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Why is the memory shortage so critical for AI development?

High-bandwidth memory (HBM) is essential for training and inference in AI systems. A shortage limits capacity, increases costs, and could slow AI progress and deployment, especially at the frontier scale.

What does SK hynix’s capacity plan mean for the market?

The company’s plans to accelerate capacity are a positive development, but the new facilities will not impact supply until 2027, leaving a short-term gap that could influence market conditions.

How do geopolitical tensions influence memory supply?

Governments view memory access as a matter of economic security, leading to interventions that may restrict exports or favor domestic manufacturers, which can affect global supply chains.

Could alternative memory architectures mitigate this shortage?

While alternative architectures may reduce reliance on high-bandwidth memory, they do not fully address the current supply constraints for large-scale AI training, which remains dependent on HBM.

Source: ThorstenMeyerAI.com

FALL

Fall Picks

As an affiliate, we earn on qualifying purchases.

You May Also Like
Fair-value Appraisals For Used GPUs And AI Hardware

Fair-value Appraisals For Used GPUs And AI Hardware

A new manual valuation method for used AI hardware aims to establish transparent market prices, benefiting brokers and resellers.
Where The 176GB Actually Goes: The Memory Budget Nobody Reads Until It’s Too Late

Where The 176GB Actually Goes: The Memory Budget Nobody Reads Until It’s Too Late

Exploring where the 176GB of model weights actually go and why total memory use exceeds initial expectations for large language models.
Review response quality coach for local service businesses

Review response quality coach for local service businesses

A new review response quality coach for local service businesses is being tested as a workflow to improve reply quality, professionalism, and compliance.
NicheCommand: A Firehose Becomes A Shortlist

NicheCommand: A Firehose Becomes A Shortlist

NicheCommand automates filtering, enriching, and scoring expired domains, turning overwhelming drop lists into actionable shortlists for buyers.