NEWS
News
We have renewed our website, which is now available in both Japanese and English. Use the “日本語” switch at the top of each page to read it in Japanese. For an introduction to ALIGN, see “About ALIGN.” Existing page addresses continue to work.
A workshop proposed by the Intelligence Symbiosis Chapter was accepted for AAAI-27 (February 2027, Montreal). It is the third workshop in the series, following those at AAAI-25 and AAAI-26.
ALIGN published the third edition (v3.1.0) of the National AGI Infrastructure (NAGI) proposal (in Japanese).
The Intelligence Symbiosis Chapter held two Proposed Sessions at the 2026 Annual Conference of the Japanese Society for Artificial Intelligence (JSAI).
The Intelligence Symbiosis Chapter released a preprint of the position paper “Humanity in the Circle (HITC): A Position Paper (in Japanese).” [Zenodo]
ALIGN set up the web page for the National AGI Infrastructure (NAGI) proposal.
ALIGN published the blog post On the Release of the NAGI Proposal (in Japanese).
ALIGN published a report on the workshop on Emergent Machine Ethics held at AAAI-26. [Japanese version] [English version]
Intelligence Symbiosis Essay 1, Does an AI Society Need an Immune System? The Significance of an AI Immune System (AIS) in Light of Yampolskiy’s Impossibility Theorem (in Japanese), was published. The LessWrong post is here.
The article Yamakawa Presents the “AI Immune System” Concept at the UAE–Japan Symposium — Proposing a Japan–UAE Joint Project (in Japanese) was published.
Thermodynamic limits of physical AI: the paper “Thermodynamic Limits of Physical Intelligence” was released. The blog post is here (in Japanese), and the LessWrong post is here.
Hiroshi Yamakawa (ALIGN, the University of Tokyo) and co-authors published “Emergent Machine Ethics: A Foundational Research Framework for the Intelligence Symbiosis Paradigm” on LessWrong.
“The Boomerang Paradox of AI Control: A Dynamic Theory of Control Decay and Time-Buying Strategies” by Hiroshi Yamakawa (ALIGN, the University of Tokyo) was presented at the 31st Meeting of the Special Interest Group on Artificial General Intelligence (SIG-AGI). (Slides)
ALIGN published the blog post Japan’s Role in the AGI Era (The Theory of AGI, AGI and Society, and Japan’s Choice: Part 3) (in Japanese).
ALIGN published the blog post AGI and Concentration Risk beyond the Scaling Race (The Theory of AGI, AGI and Society, and Japan’s Choice: Part 2) (in Japanese).
ALIGN published the blog post Can Transformers Reach AGI? (The Theory of AGI, AGI and Society, and Japan’s Choice: Part 1) (in Japanese).
“Lowering Collective Free Energy for Consensus in LLM Agent Collectives: Common Ground via Collective Predictive Coding,” by Yusuke Hayashi (ALIGN, Humanity Brain), was presented at ALife 2025.
“Universal AI maximizes Variational Empowerment,” by Yusuke Hayashi (ALIGN, Humanity Brain) and Koichi Takahashi (RIKEN), was presented at AGI 2025.
The AI Alignment Network (ALIGN) is recruiting administrative staff (a candidate for head of the secretariat). ALIGN welcomes applications from people who support its work toward “a hopeful future for AI and humanity” and want to help the community grow. Details are available here (in Japanese).
“Collective predictive coding as model of science: formalizing scientific activities towards generative science,” co-authored by Yusuke Hayashi (ALIGN, Humanity Brain), was published in Royal Society Open Science.
ALIGN published the AAAI-25 Report “Frontiers of AI Alignment” in two parts: Part 1: AI Alignment Research at AAAI-25 and Part 2: 1st Workshop on Post-Singularity Symbiosis (PSS 2025).
The “Scenario Contest on a Future Society with Superintelligence,” organized by ALIGN, received the AI ELSI Award (Special Category) of the Japanese Society for Artificial Intelligence (JSAI).
The Proposed Session “AGI Safety and Alignment for a Hopeful Future” was scheduled to be held at JSAI2025 @ Osaka, starting at 9:00 on May 29.
Koichi Takahashi (RIKEN) gave an invited talk on “The Controllability of AGI” at the 29th Meeting of the Special Interest Group on Artificial General Intelligence (SIG-AGI).
Hiroshi Yamakawa (ALIGN, the University of Tokyo) presented “A Proposal for the NAIA Vision” at the 29th Meeting of the Special Interest Group on Artificial General Intelligence (SIG-AGI).
The Post-Singularity Symbiosis WS (PSS-2025) was held and concluded successfully. (Introduction by the organizer, Hiroshi Yamakawa)
Hiroshi Yamakawa (ALIGN, the University of Tokyo) published “Reframing AI Safety Through the Lens of Identity Maintenance Framework” on LessWrong.
Hiroshi Yamakawa (ALIGN, the University of Tokyo) published on LessWrong “Proposing Human Survival Strategy based on the NAIA Vision: Toward the Co-evolution of Diverse Intelligences.”
ALIGN was scheduled to take part as a Bronze Sponsor in TAIS2025 on Saturday, April 12, 2025 (12:00–17:00, Tokyo Midtown Tower, free admission). Researchers from Japan and abroad were expected to discuss AI safety, including AI alignment.
ALIGN published AI Alignment Network Founding Symposium Report (Part 2).
Registration opened for the 13th Webinar: Allison Duettmann (CEO, Foresight Institute), “Pioneering a Safer Future with Differential Technology Development” (January 25, 10:00).
Registration opened for the 14th Webinar: Vanessa Kosoy, “infra-Bayesian physicalism and its applications” (January 27, 21:00).
ALIGN published AI Alignment Network Founding Symposium Report (Part 1).
ALIGN was scheduled to hold the “AI Alignment Network (ALIGN) Founding Symposium” on September 9, 2024. The provisional program is available here.
The AI Alignment Network (ALIGN) is recruiting volunteer staff. ALIGN welcomes applications from people who support its work toward “a hopeful future for AI and humanity” and want to help the community grow. Details are available here.
ALIGN was scheduled to take part as a sponsor in DeSci Tokyo/Funding the Commons Tokyo on July 24–25, 2024. AI alignment was expected to be one of the themes of Funding the Commons Tokyo, a forum on new forms of public goods.
ALIGN was scheduled to take part as a sponsor in the WCCI public forum “Open Forum on AI Governance” on June 30, 2024, where ALIGN’s chair, Koichi Takahashi, was to give a talk in “Session 1: Future prospects and dangers of AI.”
An award ceremony for some of the outstanding entries in the “Scenario Contest on a Future Society with Superintelligence,” organized by ALIGN, was scheduled for May 31, 2024, as part of a Proposed Session of the Japanese Society for Artificial Intelligence (JSAI).