On October 29, 2025, Anthropic opened its first Asia–Pacific office in Tokyo, marking a step in the company's international expansion. CEO and co‑founder Dario Amodei traveled to Tokyo, where he met Prime Minister Takaichi, addressed members of the Liberal Democratic Party (LDP) Digitization Headquarters Committee, met customers, and signed a Memorandum of Cooperation with the Japan AI Safety Institute. These meetings and agreements are intended to deepen Anthropic’s partnerships with the Japanese government, companies, and cultural institutions.
Building shared standards for AI evaluation
Anthropic emphasized that AI development crosses national borders and that stronger systems require international cooperation on evaluation standards: shared approaches to assess capabilities, test systems, and understand risks. As part of this effort, Anthropic signed a Memorandum of Cooperation with the Japan AI Safety Institute to collaborate on AI evaluation methodologies and to monitor emerging trends in the field.
The agreement builds on Anthropic’s existing work with AI safety organizations worldwide, including a formal agreement with the US Center for AI Standards and Innovation (CAISI) and ongoing collaboration with the UK’s AI Security Institute. In November 2024, those US and UK institutes carried out a joint evaluation of Claude 3.5 Sonnet, demonstrating how international organizations can advance AI-evaluation science together. This week Anthropic also joined the Hiroshima AI Process Friends Group, reinforcing its commitment to the 2023 framework promoting safe, secure, and trustworthy AI development while supporting innovation.
Japan’s approach to AI adoption and corporate examples
Hidetoshi Tojo, Representative Director and President of Anthropic in Japan, said that Japan’s experience supports Anthropic’s view that the most successful AI deployments augment human capabilities rather than replace them. According to Anthropic’s Economic Index, Japan ranks in the top 25% globally for AI adoption.
Anthropic highlighted several Japanese enterprises already using Claude with measurable effects:
- Rakuten uses Claude for autonomous coding projects, substantially increasing developer productivity.
- Nomura Research Institute has shortened document-analysis workflows from hours to minutes while maintaining precision.
- Panasonic has integrated Claude across business operations and consumer applications.
- Classmethod, a leading cloud integrator, reports achieving 10x productivity gains, with Claude Code generating 99% of a recent project’s codebase.
This week Anthropic also hosted its first Builder Summit in Tokyo, meeting more than 150 startups and founders building with Claude. The company reports that its run‑rate revenue in the Asia–Pacific region has grown more than 10× over the past year.
Supporting Japan’s creative community
Anthropic announced an extended partnership with Mori Art Museum. The collaboration will include long‑term work on projects such as the upcoming exhibition Roppongi Crossing 2025: What Passes Is Time. We Are Eternal., the eighth edition of a series started in 2004 that aims to snapshot Japan’s contemporary art scene. This follows the companies’ previous collaboration on the MACHINE LOVE: Video Game, AI and Contemporary Art exhibition.
Looking ahead
Anthropic stated that the people and organizations it met in Japan share the conviction that technological progress must enable human progress. The company is building a team in Tokyo to work alongside partners across industry, government, and culture. In the coming months Anthropic plans to apply the same approach in Seoul and Bengaluru as it continues its Asia–Pacific expansion, aiming to support innovation across the region.
