Messages from the President

Japanese Page

Unlocking the Future through the Power of Fundamental Science
Embarking on the Second Year of Our Fifth Mid- to Long Term Plan (April 1, 2026)

Today, human society stands at a historic turning point.

While technological innovation, symbolized by the rise of generative AI, is advancing at an extraordinary pace, global challenges?among them climate change, energy and resource constraints, international conflicts, and social fragmentation?are becoming increasingly severe.

Overcoming these challenges will require humanity to work together across differences of nationality, culture, and generation. The universality of science, trust in knowledge, and the inspiration and shared empathy brought about by the discovery of new possibilities form the firm foundation for such cooperation.

With the arrival of FY2026 on April 1, RIKEN is entering the second year of our Fifth Mid- to Long Term Plan. The purpose of this plan is to connect the pursuit of outstanding fundamental science with the creation of a better society. We are seeking to ensure that discoveries born from research that challenges the unknown do not remain the joy of researchers alone but are shared with society, and that these discoveries are nurtured into hope for a future that we can share.

To that end, we will further encourage collaboration beyond disciplinary and organizational boundaries, strengthening partnerships with research institutions around the world as well as with industry and the public.

Returning to the concept of "computation"

The problems confronting humanity are all characterized by complex interwoven causes. Methodologies capable of disentangling these complexities from a higher-level perspective and exploring solutions in new dimensions are now required on a global scale.

One concept to which we wish to return today is "computation."

The etymology of computation goes back to the Latin computare. Its root, putare, means "to prune," "to arrange," or "to think." Just as one prunes a tree while imagining its future shape, the term refers to the intellectual act of removing what is unnecessary, grasping the essence within complexity, clarifying structures, and organizing understanding into a form in which judgment and necessity can be determined. With the prefix com, the meaning expands to "thinking together and arranging knowledge into a form that can be shared."

By contrast, the other English term, "calculation," derives from the words for limestone (calx) and pebbles (calculus) used in the counting boards of ancient Rome, and thus refers to the act of arranging stones to count.

In this sense, computation is not merely standardized calculation; rather it is a creative practice. It is the process of clearly objectifying the essence of a complex world as its intrinsic relational structure and elevating it into universal knowledge that can be shared and applied. It may well be said to represent the very practice of science itself?integrating experimentation, observation, and theory.

The "Science that Connects" that we are promoting at RIKEN is precisely the challenge of adding the cross-cutting axis of computation to the accumulated strengths of comprehensive fundamental science, thereby creating new knowledge and new scientific principles across disciplinary boundaries.

Evolution of our computing infrastructure and the quantum × HPC frontier

At the core of this challenge lies the development of our next-generation supercomputer FugakuNEXT. Through integrated innovation encompassing hardware, software, and algorithms, we aim with FugakuNEXT to achieve 100 times the computing performance of the current Fugaku system. This development will serve as a foundation platform supporting a wide range of research fields.

Integration with quantum technology is another key pillar of our vision for computation. We are working on the development of the domestically produced quantum computer A in collaboration with Japanese partners including Fujitsu Limited. Since November 2023, under a project led by Japan’s Ministry of Economy, Trade and Industry through NEDO, we have introduced two commercial quantum computers employing different architectures. We are currently carrying out research and development on hybrid computational methods to connect these systems with Fugaku.

In February 2025, Quantinuum’s ion-trap quantum computer Reimei began full-scale operation on our Wako campus, initiating demonstration research utilizing high-fidelity quantum computation. In June of the same year, IBM’s Quantum System Two was installed at the RIKEN Center for Computational Science in Kobe, where development of quantum?classical integrated workflows directly connected to Fugaku is now underway. In October, we launched a test-user program for quantum?HPC collaboration aimed at practical applications such as drug discovery and materials science. And in November, we finalized the architectural design for a next-generation supercomputer premised on quantum?HPC integration.

In FY2026, we are entering the phase where we can produce tangible results from these foundations. While ensuring the stable operation and expanded utilization of the quantum?HPC integrated environment, we will continue to develop error mitigation technologies and refine hybrid algorithms. Through concrete tasks such as drug discovery, materials design, and optimization problems, we are aiming to generate outcomes that contribute both to fundamental science and to industrial applications.

Through these efforts, we will accelerate preparations for the realization of next-generation computing infrastructure in which quantum and classical systems function synergistically.

Deepening Japan-U.S. collaboration and AI for Science

Generative AI?centered on large language models?is evolving at a remarkable speed and is rapidly entering into practical use. However, the linguistic data that underpin these models will eventually reach saturation. In contrast, non-linguistic data generated by scientific measurements are effectively inexhaustible and are expanding dramatically. Advances in AI technologies to open the possibility of incorporating these data will lead to the next stage of AI.

Within this context, the development of AI for Science is accelerating. This represents more than simply an improvement in research efficiency. It is bringing about a paradigm shift in the very agents, methods, and scale of knowledge creation. This transformation will extend beyond scientific research itself and lead directly to structural changes throughout the broader socioeconomic systems that utilize scientific outcomes.

How can we accurately grasp this transformation?and how can we establish a position to lead the coming era? Answering these questions is critical for Japan’s industrial strategy and science and technology policy, and we intend to contribute responsibly to this effort.

At the same time, no single nation alone can adequately respond to these changes. In November of last year, the United States announced the “Genesis Mission” through a presidential executive order, aiming to double the productivity of its scientific research. At the heart of this initiative is AI for Science, supported by a national-scale plan to build a cutting-edge computing infrastructure.

On January 27 of this year, we signed a cooperation agreement with the US Department of Energy’s Argonne National Laboratory, Fujitsu, and NVIDIA Corporation in the fields of AI and HPC. Through the agreement we aim to advance the construction and utilization of next-generation computing infrastructure, enhance system software and scientific and engineering applications, and promote the use of AI in scientific research.

Now that AI has become a powerful accelerator embedded within scientific research, the era in which industry waits for the outcomes of basic research before moving forward is gradually coming to an end. Research sites themselves are becoming part of the front line of industry, while challenges arising in industrial settings increasingly demand new questions from fundamental science.

Japan and the United States share a vision of working together to build a global “platform for science,” centered on AI for Science as a national strategic priority. RIKEN has already been promoting the integration of computational science, data science, and AI with Fugaku at its core through the TRIP (Transformative Research Innovation Platform of RIKEN platforms) initiative, producing concrete outcomes in areas such as climate science, disaster prevention, drug discovery, and materials science.

Beginning in FY2026, we will commence full-scale operation of an AI supercomputer equipped with approximately 2,000 of the latest GPUs, including some introduced in the quantum?HPC hybrid environment mentioned above. Through the phased expansion of computational resources, we aim to construct an integrated next-generation computing environment that will work in concert with FugakuNEXT and quantum computing platforms.

The challenge of integrating HPC, AI, and quantum technologies?thereby advancing the methodology of science itself?represents the global development of “Science that Connects” and embodies the philosophy of “Made with Japan.”

Global challenges and science’s responsibility

At COP30, held in Belem, Brazil in 2025, it became evident that international discussions on climate change countermeasures are shifting their emphasis from mitigation toward adaptation. While the failure to reach agreement on emissions reduction roadmaps has drawn strong concern and disappointment, strengthening resilience remains an urgent task shared by all humanity. Both mitigation and adaptation are indispensable.

Achieving both requires a scientific foundation capable of understanding the complex systems of nature and society, quantifying risks, and predicting long-term impacts. Without scientific knowledge, the legitimacy of rational and equitable policy decisions cannot be sustained.

RIKEN is promoting research that contributes to securing the sustainability of the global commons, spanning climate, environment, ecosystems, resource circulation, and disaster prevention. High-precision simulations integrating atmosphere, oceans, and land; advances in extreme weather forecasting; and disaster risk assessments at the urban scale?all rely critically on computational science.

Toward the future

A new year is dawning in which quantum science and technology, supercomputing, and AI for Science will be connected to further advance comprehensive fundamental science.

The driving force behind all of this is people.

In FY2023 we launched the RIKEN Early Career Leaders (ECL) Program, which provides substantial support for the nurturing of young researchers. The FY2025 call for applications attracted extraordinary interest, with an application ratio of approximately 70 to 1. The fact that outstanding young researchers from Japan and abroad aspire to come to RIKEN as a place to take on challenges underscores the magnitude of our responsibility.

Grounded in the universal values of science, we will deepen collaboration with academia and industry, nurture the talent that will shape the future, and mobilize our collective strength to continue tackling global-scale challenges.

I sincerely ask for your continued support and cooperation in the year ahead.

April 1, 2026
Makoto Gonokami
President, RIKEN