A growing memory chip shortage is beginning to affect the broader tech and automotive industries, driven by surging demand for artificial intelligence infrastructure. Chipmakers are now funneling production resources toward the more profitable high-bandwidth memory (HBM) used in AI servers, reducing the availability of standard server memory used in broader enterprise environments.

The strain on supply is already showing. Samsung Electronics has raised memory chip prices by as much as 60%, while companies warn of potential production delays and rising retail prices.

With 2026 expected to bring even greater AI-related demand, experts believe the pressure on global supply chains is just beginning.

For CIOs and IT leaders planning next-generation infrastructure rollouts, this introduces new risks to pricing stability, server availability, and long-term procurement strategies into the new year.

Why It Matters: Server-grade memory is becoming more expensive and harder to secure as chip industry disruption is quickly becoming an enterprise infrastructure challenge. Procurement teams may need to revisit supplier relationships, lock in multi-quarter contracts, or explore capacity alternatives in response.

• AI Servers Are Consuming a Large Share of Memory Chip Supply: High-bandwidth memory chips, necessary for powering AI processors like those designed by Nvidia, are being prioritized by chipmakers due to their higher profit margins. These chips are now in such high demand that manufacturers of traditional consumer electronics are struggling to secure the components they need. Companies like SK Hynix and Micron are focusing their production efforts on fulfilling orders from AI datacenter operators who are willing to pay premium prices, limiting the inventory available for consumer products.

• Samsung’s Price Hikes Reveal the Growing Supply Pressure: In recent weeks, Samsung raised contract prices for several DDR5 memory modules, with the price of 32GB chips jumping from $149 in September to $239 in November. Other configurations, including 16GB and 128GB chips, also saw increases between 30% and 60%. These adjustments came without a formal announcement and were first reported by sources familiar with internal pricing decisions. Samsung has not publicly commented, but analysts say the increases are a response to tightening supplies and swelling demand from AI infrastructure projects.

• Procurement Delays Are Emerging Across the Datacenter Market: Some hardware vendors and integrators are experiencing longer lead times for memory components. As a result, datacenter projects, especially those tied to AI and high-performance computing, are being delayed or repriced. This impacts enterprise planning cycles for new infrastructure investments.

• Earlier Industry Setbacks Made the Current Shortage Worse: The memory chip industry experienced a downturn in 2023 and early 2024, which led to reduced investment in new production facilities and equipment. This decline left many chipmakers unprepared to handle the sudden spike in AI-related demand. Although efforts to expand capacity are now underway, including a new chip line announced by Samsung, those facilities will take time to become operational. In the meantime, the available supply is being stretched thin across multiple sectors.

• Long-Term Agreements and Panic Buying Are Adding to the Problem: Industry insiders report that buyers are beginning to sign long-term contracts with memory vendors to lock in prices and secure future supply. Some companies are also engaging in panic buying, ordering more chips than usual to avoid shortages down the line. These behaviors are further stressing the supply chain. While this can help reduce exposure to sudden cost spikes, it limits the ability to adjust procurement plans if project needs change. Additionally, downstream brands are likely adjust pricing strategies to offset rising input costs.

Go Deeper -> AI boom is fueling a memory chip shortage that could hit cars and phones – CNBC

Exclusive: Samsung hikes memory chip prices by up to 60% as shortage worsens, sources say – Reuters

AI Boom Triggers Memory Chip Crisis, Phones and Cars at Risk – The Tech Buzz