AISI is a student-led community at Georgia Tech working to ensure AI is developed safely, running fellowships, research projects, and policy programs across technical and governance tracks.
AISI is a student-led community at Georgia Tech working to ensure AI is developed safely, running fellowships, research projects, and policy programs across technical and governance tracks.
People
Updated 05/18/26Co-Founder & Co-Director
Founder
Co-Director
Co-Director
Fellowship Lead
Community Lead
Funding Details
Updated 05/18/26- Annual Budget
- -
- Current Runway
- -
- Funding Goal
- -
- Funding Raised to Date
- -
Org Details
Updated 05/18/26The AI Safety Initiative at Georgia Tech (AISI) is a student-led organization focused on ensuring artificial intelligence is developed to the benefit of humanity. Founded in 2022, AISI operates at the intersection of technical AI safety research, AI governance, and field-building within the Georgia Tech community in Atlanta. AISI's educational pipeline is designed to make AI safety research approachable and well-supported. Their flagship program is an introductory AI safety fellowship distilled from BlueDot Impact's AGI Safety Fundamentals course, running six-week cohorts of 6-8 students each. The program has attracted over 160 applications in a single year and over 90% of AISI's current organizers and engaged members are past fellows. For participants seeking deeper technical skills, AISI runs Rapid Upskilling Cohorts based on the ARENA curriculum, using PyTorch to build practical ML engineering competence. Select fellows are invited to join capstone research projects aimed at arXiv or workshop publication. The organization has hosted AI safety tracks at Georgia Tech's flagship AI hackathon, AI ATL, with support from Anthropic and Apart Research as sponsors. AISI also launched an inaugural AI policy hackathon in collaboration with Apart Research, covering tracks including International Cooperation, Deployment and Application Regulation, and Training Data Privacy and Copyright. AISI provides free consultation services for labs, academic departments, and PhD and MS students seeking context about AI safety, help generating compelling research projects, connections to relevant AI safety funding, and support writing grant proposals. Researchers from Georgia Tech and GTRI have also piloted in-depth crisis simulations exploring national security implications of advanced AI, designed by AISI in collaboration with Model UN at Georgia Tech. The group's membership is approximately 50% undergraduates, 45% graduate students and PhDs, and 5% working professionals. Co-director Yixiong Hao conducts research on activation engineering, interpretability, and LLM safety. AISI is recognized as an official registered student organization at Georgia Tech and can be reached at board@aisi.dev.
Theory of Change
Updated 05/18/26AISI believes that training and supporting the next generation of AI safety researchers and policy professionals at a top technical university is a high-leverage intervention for reducing long-term AI risk. By providing structured education through fellowships and upskilling programs, AISI moves talented students from awareness into active AI safety research and careers. By facilitating original research projects and connecting students to funding and publication opportunities, AISI aims to grow the pipeline of competent researchers working on alignment and governance. Hosting hackathons and policy programs brings AI safety framing to the broader technical community at Georgia Tech, expanding the base of people engaging seriously with these risks. The consulting and grant-writing support services help multiply impact by enabling other researchers and departments to pursue safety-relevant work they might not have pursued independently.
Grants Received– no grants recorded
Updated 05/18/26Projects
Updated 05/18/26Seven-week seminar on the impacts of transformative AI and agendas for effective AI governance, where participants discuss policy challenges and workshop governance solutions.
Semester-long introductory fellowship where participants join weekly discussion groups on AI safety challenges and solutions, then can progress into ML upskilling groups or supervised research projects.
Campus-wide forum featuring a keynote and parallel technical and governance workshops to help students, researchers, industry stakeholders, and the public understand AI safety risks and policy debates.
Discussion
No comments yet. Be the first to share your thoughts.