ENG
Report

Memory Crisis in the Automotive Industry: How Suppliers and Automakers are Responding

0
August 21, 2026
  • The automotive industry is managing the tight market for DRAM through a mix of long-term supply agreements and a shift to next-generation memory types, such as DDR5.
  • DDR4/LPDDR4 tightness for automotive customers will extend beyond 2027. In June 2026, Micron stated that supply would remain constrained “beyond 2027”, and no supplier has given a firm end date.
  • Chinese domestic substitution, led by CXMT, could ease pressure for the country’s market over the long run.


The global automotive market is contending with several major trends – the shift to electric drivetrains, new E/E architectures, and greater levels of driver assistance and, ultimately, autonomy. On top of these are new trade barriers, stressed consumer finances and military conflicts affecting fuel prices, which impact consumer purchase decisions and, ultimately, automaker production roadmaps. Compounding these challenges is the soaring memory demand, caused by the buildout of new data centers around the world. This has caused dramatic price increases over the last year, putting additional strain on an already stressed industry.

Here, we will focus on some of the key questions facing the auto industry now: 

How long will the memory shortage last?

Micron’s public position, reiterated during its fiscal Q3 2026 earnings call in June 2026, is that memory supply will remain constrained “beyond 2027”. That view applies to memory broadly, not automotive alone, but automotive customers sit downstream of the same constrained wafer capacity.

Micron is responding in two ways. First, by expanding capacity. The company has made an investment of over $2 billion in its Manassas, Virginia, fab, which is bringing the 1α (1-alpha) DRAM node online. The company stated that it would quadruple DDR4 production. Production started in May 2026, and Micron expects qualified 1α output by the end of the year. The company’s goal is to provide a secure, domestic source of long-lifecycle DDR4 and LPDDR4 (“LP4”) memory for automotive, defense/aerospace, industrial, networking and medical device customers, the industries most exposed to legacy-node memory being deprioritized in favor of AI-driven demand.

Second, the company has established several long-term contracts. Micron has signed Strategic Customer Agreements with suppliers including Astemo, Denso, Harman, Hyundai-Mobis, Joynext, Qualcomm and Visteon. Separately, it has signed agreements with automakers Ford and General Motors. Together, these agreements are designed to provide contracted volumes and pricing visibility for future vehicle platforms. This enhances the auto industry’s ability to plan, which is crucial given automotive production roadmaps and timelines, though it doesn’t resolve the underlying supply shortage.

Are DDR4 suppliers aligned on production plans?

The three major suppliers are not aligned on how long they will keep making automotive-grade DDR4/LPDDR4:


Supplier

Automotive DDR4/LPDDR4 stance

Stated timeline

Notes

Micron

Continuing production; expanding capacity

1α-node ramp-up underway; qualified output expected by 2026-end

Only supplier to publicly commit to ongoing DDR4/LP4 production. >$2-billion Manassas, VA, investment to quadruple DDR4 wafer supply

Samsung

Winding down, but automotive/industrial-grade reportedly on a longer runway than mobile-grade

Mobile-grade LPDDR4/4X orders already ended (April 2026); automotive/industrial-grade said to extend to Q1 2027

Q1 2027 ramp-down could be extended

SK hynix

No long-term commitment announced

DDR4 production extended through 2026

No public statement yet on automotive-specific plans beyond 2026.


Can automakers accelerate the shift to DDR5?

Automakers are being forced to shift from LPDDR4/4X to LPDDR5/Xx because (a) legacy supply is constrained and (b) newer autonomous-driving systems and platforms need the bandwidth. The newer nodes automakers are migrating to are capacity-constrained because AI/data-center demand has first claim on leading-edge output.

Examples from China of high-end autonomous-driving chips moving to LPDDR5X include XPeng's Turing (16GB x4), NIO's NX9031 (512-bit LPDDR5X) and Li Auto's M100 (16GB x4).

Micron, Samsung and SK hynix each offer automotive-grade versions of LPDDR5X. This type of memory provides a trade-off between capacity and power efficiency, while providing increased bandwidth needed to handle complex driving scenarios. As a result, Chinese EV startups have widely adopted LPDDR5X as their primary DRAM solution for L3/L4 autonomous driving.

A separate, longer-horizon dynamic is whether China’s automakers can use memory from a domestic supplier. CXMT has been advancing its automotive-grade LPDDR5 certification and BYD has taken an equity stake in CXMT. This is now translating into CXMT memory appearing in BYD production vehicles and could indicate a longer-term breakthrough. Adding memory capacity in the automotive industry’s largest market would definitely help, but how much will ultimately depend on CXMT’s capacity.

Do elevated memory prices cause problems for the auto industry?

In consumer markets like smartphones, we are seeing significant jumps in retail prices as OEMs pass along the extra bill-of-materials (BoM) costs caused by higher memory prices. However, in a smartphone, memory chips constitute the largest portion of the BoM. In a car, memory is a fraction of the BoM, so even with a 300% YoY increase in DRAM cost, the resulting impact on overall vehicle cost is not huge, though unwelcome.

Long-term supply agreements made between memory suppliers and either OEMs or their tier suppliers can mitigate price volatility, but the use of compensation pricing prevalent in other memory markets likely means the auto industry remains exposed to current and future elevated pricing.

Category

Industry

Automotive, Semiconductors

Service

Standard

Report Type

Report

Time Period

Other

Receive our insightful weekly newsletter and stay ahead of the competition.

Author

Kevin Li

linkedin_icon

Kevin is an Associate Director at Counterpoint Research based in Beijing. At Counterpoint, he leads the China automotive market research. Kevin has 12 years of experience in 5G/V2X, connected vehicles, intelligent cockpits, and intelligent driving in market analysis firms, including Strategy Analytics and TechInsights. Previously, Kevin has worked for China Unicom/China Netcom as a Senior Engineer and International Cooperation Coordinator for 10 years. Kevin holds an MSc in Mobile Communications from Beijing University of Posts and Telecommunications.

Greg Basich

linkedin_icon

As Associate Director at Counterpoint Research, Greg Basich focuses on the automotive industry. Prior to joining Counterpoint Research, Greg worked at TechInsights and Strategy Analytics for a combined 11 years, providing market intelligence and advisory services to companies in the automotive electronics market and related verticals. Prior to his career as an analyst, Greg was a business-to-business journalist at Bobit Business Media for 13 years, focused on automotive interiors, aftermarket electronics, and the fleet market. Greg holds a bachelor’s degree in economics from the University of California, San Diego.