MATS: ML Alignment & Theory Scholars

A 12-week mentored research program in AI alignment, interpretability, security and governance, with research managers, seminars and workshops. Strong fellows can apply for a funded 6–12 month extension. International applicants are supported.

• Typical window: two cohorts a year (winter and summer); register interest on the official site for upcoming dates

Official page: www.matsprogram.org/