AI safety is the work of keeping advanced AI systems from causing catastrophic harm. The field studies how those systems can pursue the wrong goals, mislead the people overseeing them, gain autonomy nobody intended, or be used to do deliberate damage. It then builds the technical methods and the policy needed to hold up against those risks.
Get Involved.
The fellowship takes applications once a semester. Everything else is open to anyone, with no application and no prerequisites. Join the Discord to hear about meetings and events.
General Meetings
One hour, every other week. You don't need to prepare anything. We cover recent AI news, discuss one topic, and serve dinner.
Fellowship
One intro fellowship in two tracks, technical and governance. Eight weeks, 90 minutes a week, and the reading happens in the session.
Speaker Events
Researchers and faculty talk about their own work. Speakers come from Texas A&M and from outside it, in person and remotely.
Socials & Dinners
We run socials through the semester, and food follows most of what we do. At the end of each semester we hold a larger dinner for everyone.
Workshops
Hands-on sessions on practical skills. Before external program deadlines, we run sessions where you work on your applications.
Reading Group
We read and break down new papers, and you can present your own research. Aimed at graduate students and advanced undergraduates, and the reading is multi-disciplinary rather than only technical.
Upcoming events
Fall 2026. Everything we run is open to anyone, with no application and no prerequisites.
Nothing scheduled right now.
The rest of the fall schedule goes up as it is set.