About ALIGN

AI Alignment Network (ALIGN) is an independent, nonprofit research network based in Tokyo, founded in September 2023 by AI researchers. It is building an ecosystem of researchers and practitioners in AI alignment in Japan and the surrounding region, and aims to serve as a hub for the collaboration across disciplines and sectors that a hopeful future for AI and humanity requires.

Our work has three parts: research; community building, including a network of the stakeholders concerned; and outreach, which covers dialogue with society and policy proposals.

Outreach: dialogue and policy

  • Since 2015. Our chair, Koichi Takahashi, co-founded the AI and Society study group in 2015 with the economist Tomohiro Inoue. It met 39 times by 2019, with researchers, executives, officials and lawyers taking part, and resumed in 2026 as the AGI and Society study group (in Japanese). ALIGN is its contact point.
  • 2026. ALIGN published a proposal (in Japanese) for shared national computing infrastructure, open to all developers and capable of supporting frontier-scale AI training. It argues that a plurality of developers would be safer than concentrating AI development in a few hands.
  • Fiscal 2025. Our chair and director Hiroshi Yamakawa served on the AI Governance Review Committee of Japan's Ministry of Internal Affairs and Communications. In addition to such formal roles, our chair holds regular informal exchanges on AI policy with senior policymakers in Japan's government and legislature.
  • 2024–2025. A public scenario contest on a future society with superintelligence, with 60 entries. The awards were presented at the 2024 annual conference of the Japanese Society for Artificial Intelligence (JSAI), and the contest received the JSAI AI ELSI Award (special category) in 2025.
  • September 2024. Founding symposium in Tokyo, with about 130 participants in person and online. Speakers included Masaaki Taira, member of the House of Representatives; Akiko Murakami, Executive Director of the Japan AI Safety Institute (recorded address); and Satoshi Kurihara, then president of JSAI.
  • June 2024. ALIGN sponsored the Open Forum on AI Governance at IEEE WCCI 2024 in Yokohama. Our chair and co-founder Ryota Kanai were among its organizers, and our chair gave a talk.
  • March 2024. ALIGN translated Yoshua Bengio's Tokyo lecture into Japanese for the Japan Deep Learning Association, which published it in two parts (part 1, part 2).

Research

Workshops and conference sessions

  • February 2027. A workshop proposed by ALIGN's Intelligence Symbiosis Chapter was accepted for AAAI-27 in Montreal.
  • 2025–2026. ALIGN and its Intelligence Symbiosis Chapter organized sessions at the JSAI annual conference in 2025 and 2026.
  • January 2026. Workshop on machine ethics at AAAI-26 in Singapore, co-organized by ALIGN director Hiroshi Yamakawa with researchers at the universities of Manchester and Bergen and others. The proceedings appear as CEUR-WS Vol-4189.
  • March 2025. 1st Workshop on Post-Singularity Symbiosis at AAAI-25 in Philadelphia, organized by ALIGN director Hiroshi Yamakawa and colleagues with ALIGN's support, with 12 accepted papers and four invited talks.

Selected papers by members

  • Koichi Takahashi and Yusuke Hayashi. "Thermodynamic Limits of Physical Intelligence." AGI 2026. arXiv:2602.05463
  • Intelligence Symbiosis Chapter. "Humanity in the Circle (HITC): A Position Paper." Preprint, 2026. doi:10.5281/zenodo.20176233
  • Yusuke Hayashi. "Lowering Collective Free Energy for Consensus in LLM Agent Collectives: Common Ground via Collective Predictive Coding." ALIFE 2025.
  • Yusuke Hayashi and Koichi Takahashi. "Universal AI Maximizes Variational Empowerment." AGI 2025, Lecture Notes in Computer Science 16057, pp. 250–262. doi:10.1007/978-3-032-00686-8_23
  • Tadahiro Taniguchi, Shiro Takagi, Jun Otsuka, Yusuke Hayashi and Hiro Taiyo Hamada. "Collective predictive coding as model of science: formalizing scientific activities towards generative science." Royal Society Open Science, 2025. doi:10.1098/rsos.241678

Essays

Community building

  • Ongoing. A Slack community of over 420 members, researchers and practitioners mostly in Japan, who share research, events and opportunities in AI alignment and safety. Anyone interested can ask to join through our contact page.
  • 2024–2025. Fourteen webinars, with speakers including Dan Hendrycks, Tan Zhi Xuan, Ryan Kidd, David Dalrymple and Vanessa Kosoy. Recordings and Japanese translations are available for many of them.
  • 2024. An introductory course on AI alignment in Japanese (four sessions), and a study group on mechanistic interpretability.

Organization

  • Registered name: 一般社団法人AIアライメントネットワーク (AI Alignment Network, General Incorporated Association)
  • Established: September 5, 2023
  • Legal form: general incorporated association (a nonprofit corporation under Japanese law)
  • Corporate number: 1010005037234
  • Office: Yamato Building 405, 1-6-16 Kanda-Izumicho, Chiyoda-ku, Tokyo
  • Chair: Koichi Takahashi. Directors: Hiroshi Yamakawa, Ippei Fujisawa, Yusuke Hayashi. See our team.

Contact

Contact us