Imagine a computer chip that thinks like a human brain but works at lightning speed using super-cooled materials. Instead of standard transistors, it uses special superconducting switches that can be triggered by light, magnets, or electricity. This allows the system to process massive amounts of data while using a tiny fraction of the power required by today's supercomputers.
Current AI and supercomputing hardware consume unsustainable amounts of energy and lack the processing speed and compactness required for advanced autonomous systems.
Fabricated YBCO/graphene junctions and tunable magnetic structures in LSMO to act as artificial neurons and synapses.
If you are an AI infrastructure provider dealing with massive electricity costs for LLM training — this project developed superconducting neurons that allow supercomputer-level processors at a fraction of the environmental cost.
If you are an autonomous vehicle manufacturer dealing with the need for real-time, low-latency decision making — this project developed ultrafast superconducting synapses that enable faster and more compact neuromorphic computing.
If you are a medical equipment manufacturer dealing with the need for high-sensitivity sensors and fast data processing — this project developed devices with combined sensitivity to light, magnetic and electric fields.
Based on available project data, specific pricing is not provided, but the goal is to reduce the environmental and energy costs of supercomputer-level processing.
The project is currently in the fabrication and testing phase of individual junctions and neural networks; industrial scaling details are not yet specified in the data.
Based on available project data, there is no specific mention of licensing terms or patent filings, though the project involves 6 partners across 4 countries.
The technology requires cryogenic environments to maintain superconductivity, meaning it would integrate into specialized cooling infrastructures rather than standard room-temperature PCs.
The project is active from 2024-05-01 to 2028-04-30, with early years focused on developing the basic Josephson Junctions.
The consortium consists of 6 partners from 4 countries (DE, ES, FR, SE). It is heavily weighted toward research and academia, with 3 universities and 2 research organizations, while industry representation is low at 17% (1 partner). This suggests the project is currently focused on fundamental technical breakthroughs rather than immediate commercial productization.
Contact CNRS in France for technical details on YBCO/graphene junction fabrication.
Contact SciTransfer to identify potential industrial partners for the 2028 transition to pilot phase.
