In a landmark collaboration poised to redefine the scientific research landscape, Japan and the United States are expected to jointly develop a specialized generative AI technology to enhance scientific research. Anticipated to commence later this year, this venture will leverage the computation powers of two of the world’s strongest supercomputers: Fugaku and Aurora.
Fugaku is developed by RIKEN, a national scientific research institute under Japan’s Ministry of Education, Culture, Sports, Science and Technology. In tandem, Aurora is overseen by the U.S. Energy Department. These two institutions, in addition to sharing supercomputing resources, will also collaborate on pooling existing and acquired data to train the AI model.
This bilateral effort is a testament to a commitment to foster “an open and enabling environment where safe, secure and trustworthy AI systems are designed, developed, and deployed,” according to a joint statement issued at the 14th U.S.- Japan Dialogue on Digital Economy.
While the formal agreement is slated to come into effect in the fall of 2024, both nations have already embarked on individual initiatives to enhance their research capabilities through AI.
In its 2023 fiscal budget, Japan earmarked approximately $79 million for AI development with a primary focus on life sciences, medical science, and material development. Additional funding is expected in the 2024 fiscal budget.
Further, Japan aims to democratize access to AI technology for corporations and educational institutions on a trial basis by 2025, with the hope of enhancing the country’s technological sophistication and international competitiveness.
Meanwhile, U.S.-backed Argonne National Laboratory is developing a large-scale generative AI technology, for scientific research in areas such as climate, cancer, and space. This complements efforts by American companies like Open AI and Google.
Both countries share a common objective: to streamline data analysis through AI.
RIKEN estimates that integrating AI into scientific research could reduce drug discovery timelines from years to a mere two months while expanding the breadth of research endeavors by almost a thousand times
Recognizing the pivotal role of high-quality data in shaping the efficacy of AI models, Japan and the U.S. have committed to sharing data and research findings. This collaborative approach ensures the development of accurate AI models that have the potential to revolutionize scientific inquiry worldwide.
