Safe AI @ Penn

Welcome to Safe AI @ Penn

Safe AI @ Penn is a vibrant, student-run group dedicated to mitigating risks associated with advanced artificial intelligence systems. These include but are not limited to: robustness, explainability, alignment, systemic safety, and monitoring. Our mission revolves around two core goals: enhancing awareness of AI safety, and providing a platform for in-depth engagement in this field through technical research.

To get involved, fill out the application form, where you can get access to our Slack and apply for the Core Research Team.

What We Do

• Weekly Reading and Discussion Groups: Delve into the latest research papers in a collaborative and insightful setting. Our sessions are split into:

  • Informal Reading Group: We read a selected paper independently and then come together for a vibrant discussion.

  • Structured Reading Group: This group follows a curriculum developed by researchers from organizations like OpenAI and DeepMind, and now adapted by organizers of Safe AI @ Penn. More information on these reading groups, as well as the full curriculum, can be found on our website: https://www.pennsafeai.org/course. 

• Speaker Events & Hackathons: Engage with experts through our speaker series and participate in hackathons to apply your skills in AI safety. We plan on hosting one public speaker event in Spring 2024, and small group conversations with researchers for members of the Core Research Team. 

• Research Opportunities: Explore pathways to work on research projects, either within the club or through external collaborations with professors and internships. We forward most external opportunities sent to us to the Slack group. 

Join Us

• Open Membership: Interested in AI safety? Fill out our application form. Everyone is welcome, and by joining, you'll be part of our community and email list.

• Core Research Team: We highly encourage people to apply to our Core Research Team for access to research support, a tight-knit research community, exclusive small-group conversations with researchers from OpenAI, and more. Apply at the link above! 

Please Note: While we are thrilled to see strong interest in the specialized team, please understand that due to the intensive nature of the work and limited resources, we may not be able to accommodate all applicants. However, this should not discourage you from applying! Every application will be thoughtfully considered, and all members, regardless of selection, will still be integral parts of the Safe AI @ Penn community with access to our general meetings, reading groups, and events.