Trust in AI deep dive: Sovereign

Technology introduction

Multi-AI agent technology enables multiple AI agents to work together as a team while continuously and safely learning from execution results, human feedback, policy updates, and specification changes. The agents identify why actions succeed or fail, extract knowledge and best practices, and apply validated learnings to future tasks.

Why it matters

This technology reduces the need for ongoing expert-led tuning by enabling AI to continuously evolve within business operations. Organizations can build and improve AI tailored to their business needs without heavy dependence on specialized AI expertise.

Example use case

In business systems such as electronic medical records, identifying the impact of regulatory or policy changes often depends on experienced specialists who know where to look across large volumes of documentation. By applying Multi-AI Agent Technology, AI agents learn from previous search results and human feedback to identify not only directly related documents but also relevant supporting information. This reduces manual effort while enabling more comprehensive and reliable impact analysis.

Technology introduction

Multiple AI systems can collaborate across attack, defense, and validation functions to identify vulnerabilities, analyze potential attack patterns, assess countermeasure effectiveness, and address emerging threats such as prompt injection attacks. Data and AI Space Collaboration Technology enables organizations to securely collaborate across company and industry boundaries while preserving data sovereignty. By combining data spaces—which allow data to remain under the control of its owner—with distributed AI execution, organizations can analyze and utilize data across multiple enterprises without moving the data itself.

Why it matters

The value of this technology is that it reduces anxiety when introducing AI into business operations, making it easier to implement and operate with confidence. As the use of AI expands, countermeasures against malfunctions, vulnerabilities, and malicious inputs become increasingly important, but it is not easy to manage these issues solely through manual efforts.By leveraging AI security technology, it becomes easier to comprehensively identify vulnerabilities and quickly implement the necessary countermeasures. As a result, even in environments with limited security experts, organizations can more easily proceed with AI adoption while mitigating risks.

Technology introduction

Data and AI Space Integration Technology enables organizations to securely collaborate across company and industry boundaries while preserving data sovereignty. By combining data spaces—which allow data to remain under the control of its owner—with distributed AI execution, organizations can analyze and utilize data across multiple enterprises without moving the data itself.

Why it matters

By securely sharing and utilizing data across organizational boundaries, enterprises can generate new insights and make more informed decisions than would be possible in isolation. This supports supply chain optimization, more accurate forecasting, and the creation of new services and ecosystems through trusted collaboration.

Technology introduction

Takane is a large-scale language model designed specifically for enterprise use, with advanced Japanese language processing capabilities and secure deployment options. While general-purpose generative AI is widely used, enterprises often need to handle highly confidential information and require a trusted operating environment.
To meet these requirements, Takane is provided in a form that can be easily used in private environments and fine-tuned to specific business operations and industries. Because it can be easily optimized using a company’s own trusted operational data, it is particularly well-suited to adapting to company-specific contexts that are difficult for general-purpose models to address.

Why it matters

This technolog makes it easier to utilize generative AI in practical business operations while maintaining confidentiality. When companies adopt generative AI, factors such as information handling, security, and suitability for their operations are just as important as performance. Takane’s value stems from its high Japanese language processing capabilities, its ease of fine-tuning with enterprise-specific data, and its design that allows for seamless integration into existing workflows and core systems. As a result, it is easy to move beyond a proof of concept (PoC) and deploy AI in actual business operations.

Example use case

In the process of policymaking and institutional design, it is essential to carefully review the large volume of opinions submitted by citizens and reflect them in policy decisions. However, due to the sheer workload involved, there is often insufficient time to thoroughly examine all feedback and incorporate it appropriately. By leveraging Takane for public comment processing, tasks such as classifying opinions by support or opposition, generating summaries, and identifying their relevance to draft legislation can be supported. As a result, administrative staff can reduce the burden of organizing public input and focus more on closely examining the content of opinions and making informed policy judgments.

Technology introduction

FUJITSU-MONAKA*1 is a next-generation, Japan-made CPU based on the Arm architecture. By combining leading-edge 2nm process technology, a 3D many-core architecture, and ultra-low voltage operation technology, FUJITSU-MONAKA is designed to deliver both high processing performance and outstanding power efficiency—approximately twice that of competing CPUs*2. It also adopts Arm SVE2 and PCI Express 6.0 (CXL 3.0), and supports Confidential Computing, helping provide the performance, efficiency, and reliability required for AI, cloud, and enterprise infrastructure. Toward the release of FUJITSU-MONAKA in 2027, Fujitsu plans to begin trials using PoC machines equipped with the new CPU from summer 2026. Fujitsu also plans to release FUJITSU-MONAKA-X, a further successor product, in 2029.

*1 This is based on results obtained from a project subsidized by the New Energy and Industrial Technology Development Organization (NEDO).
*2 Based on Fujitsu's estimated performance projections for products planned for release in 2027. Actual performance may vary depending on usage, configuration, and other factors.

Why it matters

As AI adoption expands, data centers are facing growing constraints around power capacity, cooling facilities, and installation space. FUJITSU-MONAKA applies Fujitsu’s proprietary design technologies, cultivated through the development of world-class supercomputers, to support data processing and AI inference. For mid-sized LLMs of up to around 70 billion parameters, FUJITSU-MONAKA can operate independently, contributing to building AI inference infrastructure, optimizing Total Cost of Ownership (TCO), and strengthening sovereign AI infrastructure.

Example use case

[For data centers] FUJITSU-MONAKA can be used as a next-generation AI inference platform that does not rely solely on GPUs, thanks to its design and technologies optimized for AI inference. With its high energy efficiency, it is easier to deploy in existing data centers and can support AI use cases such as RAG, Q&A, summarization, and business process automation. It can also serve as a general-purpose on-premises infrastructure platform, contributing to TCO optimization through reduced power consumption.

[For national security and highly sensitive domains] By combining the reliability of a Japan-made CPU with Confidential Computing, FUJITSU-MONAKA can be used as a foundation for sovereign AI and secure data processing in highly sensitive areas such as defense and government. It supports AI inference in closed-network and on-premises environments, as well as the modernization of infrastructure that handles confidential data.

Technology introduction

Quantum computing is a technology that performs computations by utilizing the principles of quantum mechanics, which describe microscopic phenomena such as atoms and electrons. Unlike conventional computers that process information as 0s and 1s, quantum computers use quantum bits (qubits) and leverage uniquely quantum properties such as superposition and entanglement. By using multiple states simultaneously, quantum computers have the potential to process certain types of problems far more efficiently than conventional computers.

Why it matters

The value of quantum computing lies in its potential to solve certain problems that would require an impractical amount of time on conventional computers. Applications are expected in areas such as drug discovery, catalyst development, next-generation battery materials, logistics optimization, production scheduling, and financial portfolio optimization.

Technology introduction

Quantum-HPC hybrid computing combines quantum computers, conventional high-performance computing (HPC), quantum simulators, and AI technologies to take advantage of the strengths of each computing resource. Even when practical quantum computers become available, conventional computing will continue to play an important complementary role. By integrating diverse computing technologies, it becomes possible to achieve the most suitable computational approach for a given problem.

Why it matters

In the near term, this technology enables the development and validation of quantum applications by utilizing HPC and quantum simulators, even before quantum computers can deliver sufficient accuracy on their own. Over the medium to long term, as quantum computing matures, combining quantum computing, HPC, and AI is expected to create new approaches for solving computational challenges that are difficult to address using conventional technologies alone.