Development of the Fujitsu AI Platform has entered a major phase as Fujitsu introduced a new system designed for organizations seeking full control over their generative AI operations. The company announced the launch in Kawasaki and stated that the platform will support the entire AI lifecycle within a secure, dedicated environment. This move underscores Fujitsu’s commitment to helping enterprises manage generative AI with enhanced safety, greater autonomy, and improved long-term efficiency.
With the AI platform market expanding globally, organizations are seeking greater control over their AI systems. Fujitsu explained that teams using generative AI now need more control as data protection requirements continue to rise. Many companies prefer to run their AI models internally, and the Fujitsu AI platform aims to meet this need. It allows organizations to build, run, update, and improve their models and AI agents without relying on external systems. Moreover, the company plans to roll out the platform in Japan and Europe first, with early trial registration beginning in February 2026.
The growing use of AI has also increased concerns about staffing needs, security risks, and rising computing costs. Fujitsu noted that companies often struggle with these issues, but the Fujitsu AI platform simplifies this by offering all essential tools in one package. As a result, enterprises can adopt advanced AI faster, even if they lack specialized teams or deep technical skills. The platform includes infrastructure that supports closed, private environments. It prevents external data exposure and gives customers the choice to install the system in their own data centers or inside Fujitsu facilities. This flexibility gives IT teams more control and also reduces operational pressure.
Fujitsu will support clients through deployment, setup, and ongoing management, which makes the transition smoother. Security remains a core feature of the new system. Fujitsu equipped the platform with technology that scans for more than 7,700 vulnerabilities and detects harmful attempts such as prompt manipulation and unsafe outputs. These guardrails act before and during AI execution, which helps non‑technical users maintain safe operations. Furthermore, Fujitsu plans to strengthen features that reduce hallucinations, improving trust in generated content.
Another major element is the use of Fujitsu’s high‑precision large language model, Takane. The model performs well in Japanese and includes image analysis abilities. Through the Fujitsu AI platform, companies can fine‑tune Takane to match their own business needs. The model can also be made lighter, reducing memory use by up to 94 percent through specialized optimization. This improvement lowers both cost and resource demand, making advanced AI more accessible. To support faster development, Fujitsu added low‑code and no‑code tools for building AI agents. This means on‑site teams can design and customize agents without writing complex code. The platform supports the Model Context Protocol and communication between agents, which helps them work together for more advanced applications. Fujitsu also shared plans to offer its technologies as containerized agents in the future.
Preliminary trials will begin in early 2026, introducing capabilities such as in-house model tuning and quantization. The full launch is planned for later in the year, with additional features rolling out progressively. Through the Fujitsu AI Platform, the company aims to support organizations of all sizes, from enterprise systems to edge environments, and help them adapt to the ongoing evolution of AI technologies. The broader surge in generative AI adoption is expected to further intensify competition and innovation in enterprise AI platforms. As industries search for safer and more reliable AI solutions, the Fujitsu AI platform positions Fujitsu as a strong player ready to guide businesses through their next phase of digital transformation.
