AtlasBase Launched New Thalia Bio-Silicon Chip
The company introduced a DNA-based storage chip designed to replace traditional hard drives and server infrastructure.
Updated on Sept. 22, 2026 in Semiconductors
Live Poll
Do you believe new storage technologies will effectively reduce the energy demands of artificial intelligence?
AtlasBase has unveiled the Thalia chip, a groundbreaking bio-silicon processor that integrates CMOS technology with synthetic DNA storage. This innovation aims to address the power and density limitations of modern data centers.
Why it matters
Traditional storage media struggle to support the massive power and density requirements of current AI workloads. Molecular storage offers a potential solution by retaining data integrity without the need for constant electrical power.
The Thalia chip features 5.6 billion synthesis sites and uses a 16-nanometer CMOS control layer alongside a 250nm bio-compatible device layer. Each storage capsule supports capacities ranging from 1 Terabyte to 50 Petabytes.
The players
AtlasBase
AtlasBase is a technology company specializing in molecular computing and DNA-based data storage solutions.
TSMC
TSMC is a leading global semiconductor foundry that partnered with AtlasBase to develop the Thalia chip hardware.
imec
imec is a world-renowned research and innovation hub in nanoelectronics and digital technology that collaborated on this project.
The details
The architecture utilizes a multi-layer design that combines CMOS control, a bio-compatible interface, and a fluidic enzymatic layer to synthesize DNA strands. By sending electrical pulses to individual wells, the chip triggers localized electrochemical reactions for parallel molecular computing.
Timeline
AtlasBase announced the Thalia bio-silicon chip on September 22, 2026.
The Tech Race
This development represents a departure from the limits of Linear Tape-Open (LTO) magnetic storage, which has long governed enterprise data density. By shifting toward molecular storage, AtlasBase is challenging the industry reliance on traditional magnetic and solid-state infrastructure.
For data center operators, this technology promises significantly higher storage density and reduced power consumption for massive AI datasets. Early access programs are now available, offering developers the chance to experiment with molecular compute and storage services.
The takeaway
Molecular storage could redefine long-term data archival by eliminating the constant power drain associated with modern servers. Organizations dealing with massive datasets should monitor the integration of these chips into standardized molecular cloud racks.
Further reading
For more information on the evolving landscape of processing hardware, visit the Semiconductors section.
Live Poll
Do you believe new storage technologies will effectively reduce the energy demands of artificial intelligence?
