HomeTechnologyArtificial IntelligenceDNA-Powered Memory Could Cut Computing Energy Use by 100x

DNA-Powered Memory Could Cut Computing Energy Use by 100x

​A team of researchers at Penn State University has now come up with a bio-hybrid device that pairs custom-synthesised human DNA with a perovskite crystalline semiconductor to present a new form of low-power data storage and computation. Their study was published in Advanced Functional Materials in the latest edition released August 17, 2026. Such a device may in time help reduce the energy used for powering AI computing centres and huge data centres, the scientists say.

The technology addresses a growing challenge in modern computing: the enormous energy required to model and process data. An increasing number of high-performance AI workloads and the overall compute and memory demand of data centres mean that there is pressure to find a more efficient way to store and process data.

The researchers used synthetic DNA – a chemically engineered molecule arranged into short molecular sequences and a crystalline perovskite semiconductor material (used in things like solar cells and electronics). The combined pair of these materials created a memristor: a type of electronic memory that retains clues about its past electrical experience.

Such technology could potentially go toward energy-efficient AI-computing hardware; as well as data centres and even the burgeoning field of neuromorphic computing by coupling the storage and processing information within the same device.

The development highlights a broader trend in semiconductor research: instead of relying solely on conventional silicon scaling, researchers are exploring biological materials, memristors and unconventional computing architectures to address the energy and memory demands of next-generation AI systems.

ELE Times Research Desk
ELE Times Research Deskhttps://www.eletimes.ai
ELE Times provides extensive global coverage of Electronics, Technology and the Market. In addition to providing in-depth articles, ELE Times attracts the industry’s largest, qualified and highly engaged audiences, who appreciate our timely, relevant content and popular formats. ELE Times helps you build experience, drive traffic, communicate your contributions to the right audience, generate leads and market your products favourably.

Related News

Must Read

AI Training vs AI Inference: The War for Faster, Cheaper, and More Efficient AI Computing

Abhishek Pandey/Sub-Editor/ELE Times  Learn how next-generation chip architecture, new hardware,...

Microchip Achieves QML Class V Qualification for RT PolarFire FPGA for Space Applications

Microchip Technology has expanded its radiation-tolerant (RT) PolarFire FPGA...

Vishay Extends Safety-Certified MLCC Portfolio with New 305 VAC Capacitors

Vishay Intertechnology has expanded its VJ Safety Certified Capacitors...

Infineon Partners with Megmeet to Develop 800 VDC Power Solutions for AI Data Centres

Infineon Technologies is working with Shenzhen Megmeet Electrical Co.,...

STMicroelectronics Unveils STM32G5 Microcontrollers for AI Data Centres, Robotics and Real-Time Power Control

STMicroelectronics, a global semiconductor company, has introduced its new...

AutoEV Bharat 2026, Bengaluru: A Must-Visit Show for Automotive OEMs

From intelligent vehicles and automotive semiconductors to advanced batteries...

ASMPT Introduces SIPLACE V L to Expand SMT Placement Platform Capabilities

ASMPT SMT Solutions has expanded its SIPLACE V placement...