Technical & Policy Workshops, Spring 2025
This April, the AI Safety Student Team and MIT AI Alignment had the incredible opportunity to organize three immersive weekend workshops - two focused on technical AI safety and one on AI policy - bringing together around 150 students and 20 speakers.
Set in a beautiful retreat location about an hour outside Cambridge, these workshops created a space for participants to disconnect from daily routines, engage in deep thinking, and build community through shared experiences - from intensive technical discussions to singing around the campfire.
We're grateful to all the students who joined us from Harvard, MIT, Brown, Columbia, and many other institutions. Special thanks to our exceptional speakers from Anthropic, OpenAI, METR, RAND, and The University of Texas at Austin, along with Harvard University and many others, who shared valuable insights and catalyzed meaningful discussions.
Here's to shaping a future where AI empowers us all, and doing it with care!
Participants engaging in AI safety discussions
Team picture from one of our technical workshops
Evening discussions and connections around the campfire
Student organizer giving the introductory speech at a workshop
Workshop participants taking a refreshing nature walk between sessions
Collaborative discussions on technical AI safety challenges
Building community and sharing ideas over s'mores
The beautiful retreat setting where our workshops took place
AI Technical & Policy Topics
The three workshops covered a range of critical topics in both technical AI safety and policy:
- Technical approaches to AI alignment and safety
- Governance frameworks for advanced AI systems
- Interpretability and transparency in neural networks
- Policy interventions for managing AI risks
- Collaborative approaches between industry, academia, and government
- Student-led initiatives in AI safety research and governance
Through hands-on technical sessions, expert panels, and collaborative discussions, participants developed practical skills and deeper understanding of the challenges in ensuring AI benefits humanity from both technical and policy perspectives.