SoftBank and Five Japanese Giants to Build Quantum Materials Platform

SoftBank has signed a joint research agreement with five Japanese industrial groups — AGC, Toppan Holdings, Aisin, Asahi Kasei and DIC — to build a materials development platform that combines quantum computing with AI, with funding from Japan's New Energy and Industrial Technology Development Organization (NEDO).

The project will run from fiscal year 2026, which ends March 31, 2027, through fiscal year 2028. That gives the six-company consortium roughly two to three years to develop and validate the platform. NEDO selected the proposal under its call for the "Research and Development Project of Enhanced Infrastructures for Post-5G Information and Communication Systems," placing the work inside a national program aimed at keeping Japan's industrial edge in the global market.

The partners span chemicals, automotive components, materials and printing. SoftBank will connect quantum computers and supercomputers through a unified platform to generate material-development data, build AI-based data utilization technologies, and run demonstrations that optimize material development across multiple fields.

Why combine quantum computing with AI?

In a joint statement, SoftBank framed faster materials discovery as an urgent priority. The operator pointed to two existing approaches and their limits: data-driven methods, where AI predicts and explores new materials from historical data, struggle to secure sufficient, high-quality, high-precision training data. Computational science simulations based on physical laws face surging computational costs as accuracy and scale increase.

"In recent years, progress has been made in data-driven approaches – where AI predicts and explores new materials based on historical materials data – as well as computational science simulations based on physical laws," SoftBank said. "However, while data-driven methods face a challenge in securing sufficient, high-quality, and high-precision training data, computational science faces the issue of surging computational costs for higher accuracy and larger scale calculations."

According to SoftBank, quantum computers can simulate molecular and electron behavior more efficiently than traditional machines. Present-day limits on qubit scalability and error correction, however, prevent large-scale computations on quantum hardware alone. The project's answer is a hybrid: AI-driven methods paired with quantum calculation to soften the constraints on precision and scale. The stated goal is to drastically shorten the R&D cycle from theoretical molecular design down to physical verification.

How will validation work?

The platform will target high-demand materials in three areas:

• next-generation battery materials
• optical materials for new displays and sensors
• superconducting materials for future devices

A continuous feedback loop sits at the core of the design. Materials will be physically synthesized based on quantum computing simulations, and the resulting experimental data will be fed back into the system to improve predictive software accuracy.

"Rather than targeting a single specific material, the materials development technologies will be applied cross-sectionally across a broad range of industrially vital materials systems," SoftBank said. "Each of the six companies will handle verification in their respective domains of strength." The consortium will also work with specialized quantum computing institutions to test how well the approach performs across different materials, with the aim of creating practical industrial value.

Where does this fit in SoftBank's quantum strategy?

The NEDO-backed project aligns with SoftBank's wider push to develop practical industrial applications of quantum computing in Japan. It complements the operator's existing collaboration with Quantinuum, which is building a roadmap that links quantum algorithms, hardware evolution and commercial use cases. Together, these efforts position SoftBank as a participant in moving quantum technologies from research into industrial deployment.

The commercial outcome remains distant by design: no product or market deployment is expected before the consortium completes development and validation work in fiscal year 2028, with the six companies then applying the platform in their own materials domains.

By Elena Vasquez