World’s first enterprise-wide generative AI framework technology

Fujitsu Limited proudly announces the development of the world’s first generative AI framework designed specifically for enterprises. Starting in July 2024, this innovative framework will be available globally as part of the Fujitsu Kozuchi lineup.

Key Features:
1. Knowledge Graph Extended RAG: Enhances data input accuracy by expanding data references for large language models to over ten million tokens, utilizing knowledge graphs.
2. Generative AI Amalgamation Technology: Automatically selects and combines specialized AI models to meet diverse enterprise needs, ensuring high performance and sustainability.
3. Generative AI Auditing Technology: Ensures corporate and legal regulations compliance through explainable AI and hallucination detection technologies.

Benefits:
1. Efficiency Gains: Expected to reduce contract compliance verification person-hours by 30%, improve support desk efficiency by 25%, and cut optimal driver allocation planning time by 95%.
2. Diverse Applications: From analyzing large-scale genome data to enhancing support desk operations and optimizing transportation logistics.

Fujitsu will expand its lineup of specialized AI models, focusing on applications that include Japanese language processing and coding generation. The company’s proposal for developing lightweight knowledge graphs for secure on-premise environments has been accepted by GENIAC, a Japanese Ministry of Economy, Trade and Industry project.

Fujitsu remains committed to advancing AI technology to meet enterprises’ evolving needs and provide robust support for their business operations.

Read more here

Connect me at LinkedIn