Carnegie Mellon University (CMU), Arm and SoftBank Group Corp. (SBG) announced that Arm and SBG will contribute USD 15.5 million to CMU to support its partnership with Keio University (Keio), a collaboration to accelerate the global advancement of AI.
A year ago, the United States and Japan launched a research partnership to accelerate advancements in AI on a global scale. The partnership will bring the two countries closer together as international leaders in AI by pairing up CMU with Keio and, in a related effort, the University of Washington with the University of Tsukuba. Several global companies — including Amazon, Arm, Microsoft, NVIDIA and SBG — and a consortium of Japanese companies pledged USD 110 million in contributions to support these collaborations.
Arm, the company that is building the future of computing, and SBG, a global investor in breakthrough technologies, will provide CMU with USD 15.5 million to support the university’s efforts in this partnership. The funding will give CMU scientists access to cutting-edge commercial tools and models and support fundamental research in areas where AI will have a transformative effect.
“Carnegie Mellon is thankful for the support of Arm and SoftBank Group Corp. to further the development of transformative technologies powered by artificial intelligence. AI is pushing scientific discovery in fields like robotics and biomedicine, helping researchers understand complex systems and predict outcomes with increased accuracy,” said Martial Hebert, Dean, CMU’s School of Computer Science. “The CMU and Keio partnership will unlock research potential across the globe. Together, with generous and expert industry backing, we will blaze the new path charted by pioneering breakthroughs made possible through AI.”
As a key feature of the partnership, Arm will provide CMU researchers with access to their hardware and software IP and tools through the company’s Academic Access model, allowing researchers to innovate, evaluate, design and manufacture with the latest technology from Arm. The company’s funding will also support research collaboration, capstone projects, internships and other student experiences between CMU and Keio.
“At Arm, we see the future of AI at the intersection of cutting-edge research, world-class engineering and practical application. Our partnership with the SoftBank Group and Carnegie Mellon University accelerates AI innovation while empowering the next generation,” said Khaled Benkrid, Senior Director, Education and Research, Arm. “By uniting academia and industry, we’re not just advancing technology, we’re also investing in its future architects.”
SBG will establish the SoftBank Group–Arm Fellowship with its contribution. The fellowship will support CMU’s Ph.D. students in AI research. The funding will also support student-focused academic programs, such as summer research opportunities for undergraduate students, capstone projects and internships.
“AI’s true potential emerges at the nexus of diverse ideas, rigorous scholarship and strategic international alliances,” said Tim Mackey, Corporate Officer, CLO & GCO, Head of Legal Unit of SoftBank Group Corp. “Through our support for the Carnegie Mellon–Keio initiative, we seek not only to catalyze innovation but also to strengthen the foundational bonds between Japan and the U.S., building a robust pipeline of visionary AI talent prepared for global leadership.”
With support from Arm and SBG, researchers at CMU and Keio will embark on work in four main areas: multimodal and multilingual learning, embodied AI for robotics, autonomous AI symbiosis with humans, and life sciences and AI for scientific discovery. Collaborations between CMU and Keio have begun. Researchers from Keio visited CMU in the fall of 2024. CMU researchers visited Keio in the spring of 2025.
CMU and Keio researchers in a current project in embodied AI are developing systems that will allow robots to better understand their own abilities, making them more effective at completing tasks around the home. Anticipated research in multimodal, multilingual learning will look at methods to reduce hallucination problems in large language models when they are used to analyze images or other multimodal data. Projects in AI symbiosis could include research into understanding non-verbal communication, interaction with robotics arms and social navigation. Researchers looking at the use of AI for scientific discovery will investigate AI-controlled experimentation and data analysis, including biomedical image analysis, lab automation and AI-enabled interpretation of molecular measurements.
