AI Safety Fellowship applications are now open; apply before February 6th!

A community at Georgia Tech ensuring AI is developed to the benefit of our future.

Our Mission

Managing risks from advanced artificial intelligence is one of the most important problems of our time. We are a community of technical and policy researchers at Georgia Tech aimed at reducing these risks, training the next generation of researchers, engaging the public, and steering the trajectory of AI development for the better.

Research

We pursue and publish novel work that seeks to understand, evaluate, and make safe powerful artificial intelligence systems, including through interpretability, goal specification, and governance projects.

View projects »

Recent Publications:

Patterns and Mechanisms of Contrastive Activation Engineering

2025 ICLR Workshops, Multiple

Robust Unlearning via Mechanistic Localizations

2024 ICML MechInterp Workshop Spotlight

A Mechanistic Analysis of a Transformer Trained on a Symbolic Multi-Step Reasoning Task

2024 ACL main conference paper

Education

We host a variety of educational fellowships, upskilling programs, and large events exploring the fields of AI alignment and governance, drawing in talent from around Atlanta and the Georgia Tech community.

View all programs »

Upcoming: Technical Safety Fellowship

A 6-week seminar that introduces the catastrophic risks associated with advanced AI, and current research agendas in AI Safety. Explore challenges and present solutions to robustly aligning AI systems with human intentions through weekly discussions.

Upcoming: AI Policy Fellowship

A 7-week seminar that explores the potential impact of transformative AI, and introduces agendas for effective governance. Discuss the obstacles of ensuring that advanced AI systems are developed and deployed safely and for the benefit of humanity, and workshop policy solutions.

We seek to engage a wide variety of students, faculty, and researchers from across fields and departments. Our AI Safety Pipeline is designed to make technical and governance research approachable and well-supported at Georgia Tech, and we are constantly looking for innovative and creative ways to further our mission.

Events

We host various events including open meetings, speaker events, and reading groups to foster engagement within the AI Safety community and promote the education of AI Safety issues. To stay up to date on all our events, you can view the calendar below and join our discord server.

Open Meetings

Meet with AISI and extended community at beginner-friendly gatherings, whether over eating pizza during member-led presentations, engaging in lively trivia games, or diving into journal discussions.

Speaker Events

Join us for engaging presentations by leading researchers and industry professionals who share their insights on AI safety and showcase their latest research. These events, open to both members and the public, offer a unique opportunity to explore emerging topics in the field, engage directly with speakers, and connect with others who share your interest in ensuring AI development benefits humanity. To stay up to date about upcoming speaker events, please join our discord server.

Empirical Research Reading Group

This group focuses on new exciting papers in empirical AI safety and alignment research and caters towards a graduate student audience. We meet every Friday 5-6:30PM in Coda C1115 Druid Hill. More detail here.

Keep track of our events!

This calendar will be updated with upcoming events and deadlines. Sync this to your calendar and watch out for updates!

Consult with us

We provide free consultation services for labs, academic departments, and Ph.D./M.S. students to help provide context about the field, generate compelling projects, connect projects to relevant AI safety funding, and write grant proposals. Send an email to board@aisi.dev describing your situation & the level of funding you seek and we'll be in touch!