Study Groups

Study Groups

Study groups organized by members of the ALIGN community.

[First cohort completed] Introductory Course on AI Alignment

We ran a four-session introductory course for quickly learning the technical foundations of AI alignment (July–August 2024). Participants read carefully selected materials and discussed them with experienced facilitators. The course drew on materials from AI Safety Fundamentals, which provides comprehensive learning materials for the field. Details (in Japanese).

Lead: Masayuki Nagai (Moon)

Mechanistic Interpretability Study Group

A study group on mechanistic interpretability (article in Japanese), which has drawn attention in recent years in the context of AI alignment as well. We plan weekly reading sessions, information sharing on Discord, and research activities such as hackathons. If you are interested, please let us know via the contact page.

Lead: Ryota Takatsuki @rtakatsky