London AI Safety & Alignment Forum

A forum that grew out of a small reading group that met every week to go through papers together. As AI grows powerful so quickly, we felt we needed a place to take safety and alignment seriously. Over two days we cover interpretability, evaluation methodology, and governance. Mornings are deep technical sessions; afternoons are roundtables where researchers and policy people sit around the same table. This is not a place that hands you answers, but one where we hold onto the hard questions together. Whether you come from academia, industry, or policy, anyone thinking seriously about this is welcome.