Fujitsu unveils AI agent technology that evolves by learning from business operations

Fujitsu Develops Self-Evolving Multi-AI Agent Technology

Fujitsu has announced the development of self-evolving multi-AI agent technology that enables AI agents to perform tasks as a team while continuously learning from daily execution results, human feedback, policy revisions, and specification changes 1

1

2

2

. This Adaptive AI Agent Framework addresses a critical limitation in conventional AI systems: while traditional AI agents demonstrate high processing capabilities for given instructions, they struggle to independently analyze reasons for failure and safely incorporate lessons into subsequent business operations 2

2

.

The technology represents a shift in how AI agents learn and adapt to dynamic business needs. In corporate environments, legal revisions, system changes, specification updates, and on-site rule modifications occur continuously. Previously, determining which information to reference and which judgment criteria to prioritize relied heavily on the experience and tacit knowledge of skilled professionals 2

2

. Experts needed to continuously adjust prompts, search methods, evaluation criteria, and operational rules to keep AI systems current.

Autonomous Business Execution Through Continuous Learning

The most significant feature of Fujitsu's AI agent technology is that while performing tasks, AI agents identify reasons for success and failure, extract actionable knowledge and operational insights, and verify improvement proposals before implementation . This allows AI agents to take over tasks such as prompt adjustments and evaluation criteria updates, which were previously performed continuously by experts. By deploying AI within the customer's environment, it continuously adapts to individual rules and judgment criteria that arise during business operations, creating a business foundation that evolves with people and the environment.

Fujitsu automatically enhanced "Takane" for multiple domains such as manufacturing, healthcare, finance, and public administration. Through operational use, continuous improvements resulted in a significant average accuracy improvement of 28 points compared to pre-specialization performance 2

2

. In the medical field, the technology enables structured extraction of information tailored to specific operations, such as extracting diagnostic names, progression stages, and treatment policies from unstructured data like medical records and test results.

Data and AI-Driven Management at Scale

The technology can be applied to the entire process of building business-specific LLMs (Large Language Models). Multi-AI agents autonomously execute and optimize a series of steps such as data selection, adjustment of learning conditions, evaluation, and improvement, which were previously handled by experts 2

2

. This enables companies to build AI tailored to their own operations in a short period and continuously improve it in response to changes, without heavily relying on AI specialists.

Under the OneFujitsu initiative, Fujitsu has been promoting the global standardization of internal IT, data, and business processes. By globally applying this technology and the multi-AI agent platform equipped with Takane, Fujitsu is achieving autonomous business execution and accelerating management speed, thereby accelerating the transition to data and AI-driven management 1

1

2

2

.

Real-World Application in Healthcare and Government Systems

This technology was applied to AI agent-based document search for design specifications of Fujitsu's electronic health record system for medium-to-large hospitals and business solutions for local governments [1](https://digitalterminal.in/enterprise/fujitsu-l aunch-adaptive-ai-agent-framework-for-dynamic-business-needs). Traditionally, identifying the scope of impact for software modifications due to legal revisions or policy changes required skilled experts with deep knowledge of regulations, business processes, and system architecture.

With this technology, AI agents now learn from past search results, failure cases, and human feedback. As a result, they autonomously improve search range expansion and document extraction strategies, reducing the effort required for designing and improving search logic while improving accuracy 1

1

. The AI agents demonstrated they could learn and apply exploration techniques used by skilled experts, such as checking related peripheral documents during operations and not excluding seemingly irrelevant documents if they belong to the same business domain.

Sovereign AI for Edge and On-Premises Environments

Fujitsu plans to integrate this technology into its proprietary AI platform and offer it as a core technology to support the in-house development and autonomous operation of business-specific AI 1

1

. As one of the advanced AI technologies within the Fujitsu Kozuchi AI platform, the company will promote its application to a wide range of areas requiring specialized knowledge and continuous improvement.

By combining insights from joint research with Associate Professor Graham Neubig and Assistant Professor Tim Dettmers from Carnegie Mellon University, Fujitsu will advance the development of technology to operate self-evolving multi-AI agent systems with less memory and power 1

1

. This aims to enable AI teams that continuously learn from operations not only in cloud environments but also in highly confidential on-premises and edge environments. Fujitsu aims to realize sovereign AI that can continuously learn across these diverse deployment scenarios.

Addressing Workforce Challenges and Knowledge Transfer

By enabling AI to learn from on-site failures, human instructions, and environmental changes in real-time, and safely apply this knowledge to subsequent tasks, Fujitsu will evolve AI into an intelligent foundation that grows with the workplace 1

1

. This approach addresses critical societal challenges including personnel shortages, adaptation to regulatory changes, and knowledge succession of tribal knowledge. The technology aims to create a future where people and AI learn from each other to advance entire industries.

Fujitsu's commitment extends to the Sustainable Development Goals (SDGs) adopted by the United Nations, with the company's purpose being "to make the world more sustainable by building trust in society through innovation" 1

1

. The self-evolving multi-AI agent technology represents a step toward this vision by democratizing access to specialized knowledge and creating adaptive systems that can serve organizations of varying sizes and capabilities.