Fujitsu Launches Open Source Projects for Automated Machine Learning and AI Fairness

Fujitsu Launches Open Source Projects for Automated Machine Learning and AI Fairness
Image Credit: Maginative

Fujitsu, Japan's leading developer of AI technologies, recently announced a major commitment to open source AI by launching two new projects with the Linux Foundation. The projects provide tools that can automatically generate code for new machine learning models and address latent biases in training data.

By open sourcing these key technologies, Fujitsu says it aims to spur AI innovation and address concerns around transparency and ethics. The timing is strategic, just ahead of the "Open Source Summit Europe 2023" in Bilbao, Spain, set for September 19-21.

The two technologies at the core of the new Linux Foundation projects represent important advances in AI that have broad applications across industries.

SapientML, the automated machine learning project, can rapidly generate optimized machine learning models and code tailored to a company’s specific data. This empowers data scientists and engineers to quickly build and refine precise models, accelerating development.

AutoML technology for automated code generation under development in the SapientML project

The Intersectional Fairness project provides technology to uncover and mitigate biases in AI systems, particularly subtle biases that arise at the intersection of attributes like gender, age, and ethnicity. Mitigating these overlooked biases is critical for developing fair, ethical AI.

Example of often undetected intersectional biases that can be discovered and reduced by AI fairness technology

By providing access to these technologies as open source, Fujitsu expands their availability beyond its own advanced AI platform. Developers worldwide can now innovate on top of these capabilities to create solutions for new domains.

“Offering AI technologies as open source lowers the barrier of entry for developers across industries,” comments Vivek Mahajan, CTO of Fujitsu Limited. This furthers the company’s vision of open collaboration advancing AI for societal benefit.

With over 970 AI-related patents from 2014 to October 2022, Fujitsu leads AI development in Japan. Earlier this year, the company launched "Fujitsu Kozuchi," an AI platform enabling rapid and secure testing of advanced AI technologies. This prodigious innovation output underpins the company’s strong market position in digital services and advanced technologies like AI.

By choosing to work with the Linux Foundation’s neutral, non-profit oversight, Fujitsu leverages the gold standard in open source governance. This will foster global participation in maturing the technologies.

Jim Zemlin, the Executive Director of the Linux Foundation, anticipates that these OSS contributions from Fujitsu will significantly contribute to the advancement and distribution of AI technologies. The Linux Foundation, a non-profit technology consortium, vets and approves roughly a dozen technologies as OSS projects annually, prioritizing open technologies that solve pressing issues.

As AI continues to shape the modern world, initiatives like these offer a glimpse of how collective, open-source efforts can steer the technology in a direction that aligns with societal values. It signals a future where AI serves not just a privileged few but stands as a tool for global innovation, subject to rigorous ethical standards.

Let’s stay in touch. Get the latest AI news from Maginative in your inbox.