Fujitsu and Mizuho have started a joint trial to apply Fujitsu’s generative AI technology to Mizuho’s system development process. The purpose is to verify the effectiveness of AI in automatically detecting errors and omissions in system design planning and audit processes.
The trial will run from June 2023 to March 2024.
Generative AI technology offers tremendous potential to transform the way we work. With rigorous testing, the day may soon come when companies will be able to optimize various business processes and tasks with AI.
Recognizing the potential of this technology, Mizuho is currently exploring ways to incorporate AI into its operations. The company believes AI will improve system quality and resilience during development and maintenance.
AI code name: Fujitsu Kozuchi
Fujitsu is actively promoting research and development of generative AI models. We support system development and maintenance technology within the framework of “Fujitsu Kozuchi (development code name) – Fujitsu AI Platform”. This will enable users around the world to accelerate the testing and deployment of advanced AI technologies.
On June 9th, 2023, Fujitsu began offering an AI core engine for generative AI. The generative AI technology used in the demonstration experiment is a combination of this AI core engine for generative AI and Fujitsu’s engineering experience.
The two companies also plan to develop technology to automatically generate test specifications from AI-enhanced system design plans. To strengthen the resilience of business processes, Fujitsu and Mizuho plan to apply generative AI technology to business processes and tasks. This includes source code generation, system development and maintenance.
Fujitsu is collaborating with Tokyo Institute of Technology, Tohoku University, and RIKEN. In May, they announced the start of a research and development project to conduct distributed training of a large-scale language model (LLM) on the supercomputer “Fugaku”, which is scheduled to be released in 2024.
Fujitsu will utilize the research results of this project to promote practical application research in the industrial world. By applying these to Mizuho’s systems, we will achieve further quality and efficiency improvements in Mizuho’s operations.
According to Manish Dixit, head of practice at RBI publisher GlobalData, generative AI has the potential to transform financial services.
“Generative AI in financial services can be used to develop intelligent chatbots and virtual assistants that provide personalized customer support. These AI-powered assistants can answer customer questions and perform basic transactions. Leverage large-scale language models to analyze financial data to deliver valuable insights, personalized recommendations tailored to customers’ financial goals and preferences, and insightful performance comparisons .”
