- As AI data centers consume an ever-increasing amount of RAM chips, researchers are looking for cheaper and more scalable alternatives.
- Imec’s research is seen by many as the answer, although engineering challenges continue to exist.
- Two published studies using ferroelectric materials show capacitors essentially capable of replicating modern DRAM functionality.
It’s no secret that AI has affected much of the world, with many welcoming the advent of an “intelligent” digital assistant, while others are using it to speed up the most mundane tasks that previously required monitoring.
On the other hand, some find their jobs redundant, others are forced to upskill or turn to other sectors just to earn a living in 2026.
There is, however, one area that has been significantly impacted and whose effect may have been underestimated until now: the computer hardware industry.
DRAM for data centers first
The reason why the hardware industry (and, to a lesser extent, the smartphone and gaming console industry), or any industry that uses fast RAM and NAND flash for its core functionality, is affected by the overwhelming demand from data centers.
With billions of dollars in construction and procurement budgets, the average consumer is unlikely to compete for memory with a data center built by hyperscalers and for good reason: There’s simply too much money at stake here.
The data center problem, however, is different: despite consuming nearly 70% of all memory produced in 2026, resulting in the worst memory supply crisis in 15 years, the situation will only get worse, even as demand continues to grow unabated.
While memory giant Micron is telling consumers to wait until 2028 before they can expect relief in their segment, and SSDs are following the same path, thanks to a similar situation in high-density NAND flash storage, AI data centers are also looking for a breakthrough in memory that could enable cheaper and faster memory in the near future to better meet their needs.
European chip research lab Imec, however, appears to be working on an answer. At the 2026 IEEE/JSAP Symposium on VLSI Technology and Circuits, the event showcased two major advances that could bring new ferroelectric memory into the mainstream over the next decade as it attempts to address the AI-centric memory shortage with a solution better suited to data center consumers.
At the heart of today’s advances is a ferroelectric capacitor that operates at low voltage, allowing a high number of write cycles and retaining charge well, making it a potentially viable replacement for traditional DRAM memory.
The second is a transistor that they managed to stack vertically, allowing for a denser NAND flash-like storage design while adding a rear-door modification to address its memory erasure issues.
FeRAM is not a new concept; First conceived in 1952, it shows enormous promise, although most researchers agree. Until recently, its interest was limited, but a world spending billions on AI data centers, with increasingly scarce memory and storage resources, has brought it back to the forefront, even if most of its gains still limit it to laboratory conditions.
“This work shows how Imec’s multidisciplinary expertise, from materials science to advanced 3D integration, allows us to address some of the most pressing challenges in memory technology,” said Maarten Rosmeulen, program director at Imec. “We are exploring multiple paths to the memory solutions that will be needed to support the rapid growth of AI and data-intensive applications.”
It should be noted that IMEC does not operate in a vacuum; It shares its research with hundreds of industry partners, including chip designer Nvidia, photolithography systems maker ASML and manufacturing giant TSMC.
Other industry partners include Intel, Samsung, Micron, Qualcomm, AMD and Apple, indicating that its research could potentially be integrated into future memory offerings from several interested entities, including other hyperscalers.
If Imec’s research solves the modern data center conundrum by offering a cheaper, denser memory module using different materials, it could very well spark a new AI-centric war for high-end storage and memory.
But for now, it appears to be a long time before it’s ready for production, with Imec admitting it’s “addressing remaining challenges” while reminding interested parties that this entire advancement was still a proof of concept and still in the research stage.
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds.
