Fujitsu Research News Update - June 2024

Launching Fujitsu's Enterprise-Wide Generative AI Framework

At the Fujitsu Research Strategy Briefing on 4 June, Head of Fujitsu's AI Laboratory Toshihiro Sonoda unveiled our exciting generative AI framework for enterprises, which will be available globally as part of the Fujitsu Kozuchi lineup from July 2024.

The new framework has been developed to respond flexibly to the diverse and changing needs of companies, as well as allowing for the easy compliance with the vast amount of data and laws and regulations that companies possess.

Fujitsu has developed its new generative AI framework for enterprises in order to strengthen specialized AI that is able to solve these issues for companies. The new framework consists of:

• knowledge graph extended retrieval-augmented generation (RAG), using knowledge graphs to link the relationships between large-scale data that companies possess and enhance the data input to the generative AI
• generative AI amalgamation technology, which selects the model with the highest performance from multiple specialized generative AI models based on the input task or it automatically generates by combining the models
• the world’s first generative AI auditing technology enables explainable output which will adhere to compliance with laws and company regulations.

Fujitsu is currently conducting a verification test using its generative AI framework for enterprises. It is expected to achieve a 30% reduction in man hours for contract compliance verifications, a 25% improvement in support desk work efficiency, and a 95% reduction in the time it takes to plan optimal driver allocation in the transportation industry.

Focus on AI Ethics

At Fujitsu, our approach to AI Ethics involves advocating for human-centric technology generation processes. We have been working on AI Ethics from this perspective, and accumulating know-how from an early stage – all in order to build trust in society about the safe and responsible use of AI.

Among our activities to promote the development of responsible AI Ethics, we became a founding member in 2018 of AI4People, a global consortium group of Europe with expertise in the impact of AI on society. This partnership has enabled us to formulate our AI commitment, working within AI4people's 5 principals.

We have continued to play an active part in the group's activities, most recently involving an important new report published by the AI4People Institute - Towards An Ethics By Design Approach for AI. The report represents a significant milestone in transitioning the conversation from ethical principles for AI to practical implementation.

The formal launch follows a recent Presentation held on 21 May, at which Fujitsu Research’s Virginia Ghiara participated as a co-presenter. Fujitsu, as Co-Chair of the AI4People Institute’s Scientific Committee, played a key role in shaping this approach by contributing to the development of the proposed guidelines and supporting the collection of best practices in AI ethics. The AI4People Institute’s comprehensive guide offers valuable and practical insights on designing, developing, and maintaining AI systems that align with the laws, ethical principles, and values of the European Union.

A recording of the launch event is available here: https://lnkd.in/djGR3BJe - where Fujitsu’s representative Virginia participated alongside AI ethics experts and representatives from the European Parliament, European Commission, and the Council of Europe.

On the World Stage!

Spreading the word at Linaro Connect in Madrid

The recent Linaro Connect 2024 event in Madrid, Spain provided the Fujitsu team with a great platform from which to share our latest developments in Arm AI and LLVM software development.

The Fujitsu team, led by Kouichi Hirai, Shun Kamatsuka, Ryotaro Kasuga, and Ragesh Hajela, spotlighted the upcoming revolutionary 2nm ARM-based CPU, FUJITSU-MONAKA, designed for 2x performance through energy-efficient computing, emphasizing ease of use, high reliability, and confidential computing. The response from conference attendees was phenomenal!

Focusing on Generative AI at WCRI 2024 in Athens

?Fujitsu Research’s Senior Research Manager, Dr Shanshan Yu, presented a keynote on “Trustworthy Generative AI from the perspective of industry” at the recent WCRI 2024 – the largest and most significant international conference on Research Integrity and Responsible Conduct of Research.

?Dr Yu set out how Fujitsu's technologies can support people to fulfill their specific purposes by using Generative AI technology, as well as improving the performance and trustworthiness of Generative AI?itself. Her keynote included a detailed explanation of the Fujitsu Neuro-Symbolic Explainer technology that enables the extraction of validated explanations for image classifications made by AI. It is one of the AI core engines of Fujitsu Kozuchi, which enables the fast testing of cutting-edge AI technologies developed by Fujitsu.

For more information on our technologies:

• Introducing Synthetic Image Generation
• Fujitsu launches new technologies to protect conversational AI from hallucinations and adversarial attacks?

New Technology Spotlight - Security Audit Automation

This month, we would like to introduce our new Security Audit Automation technology, representing a new frontier in security with large language models (LLM) that helps companies to to strengthen their security measures against increasingly sophisticated cyber attacks.

Security audits, which identify and remediate vulnerabilities in systems and applications, are an essential part of security measures - helping organizations to evaluate whether security measures for systems and applications are implemented properly, identifying security issues, and improving them.

For a deep dive, read the Fujitsu TechBlog!

Our new Security Audit Automation technology is available for you to test out - if you are interested, please log in to the Fujitsu Research Portal, select the "Security Audit Automation" page from the technology list, and try out the demo site. Or just click the link here to find out more.

In the News - Highlights?

Fujitsu has secured its first commercial order for a gate-based superconducting quantum computer - awarded by the National Institute of Advanced Industrial Science and Technology (AIST)

The new superconducting quantum computer system is designed to scale to hundreds of qubits, one of the largest in the world, without upgrading the dilution refrigerators that constitute a superconducting quantum computer. It applies Fujitsu’s newly developed technology for mounting high-density wiring inside refrigerators and packaging technology for large qubit chips.

Read the full press release here: Fujitsu to introduce superconducting quantum computer system at National Institute of Advanced Industrial Science and Technology

Our Team

In this latest Researcher's Dream article, Tatsuya talks about his fulfilling career in cutting-edge network technology, sharing his insights en route to his focus today on R&D aimed at enhancing networks using AI. He also talks about his passion for soccer - he's a keen member of the the Fujitsu soccer team.

If you know someone who may be interested in the Fujitsu Research News Update, please forward it to them! Thank you.

Find out more about our global research activities at Fujitsu Research and Development