Dartmouth's student organization dedicated to reducing catastrophic risks from AI.
Reducing risks from advanced AI may be one of the most important problems of our time. We run an introductory reading group on why AI safety matters and what we can do about it. Open to anyone regardless of background, undergrad or grad students.
Reading group
Monday evenings, 90 minutes/week for 6 weeks, starting week five of the term. There is no preparation required since we read together, and will have good food for dinner. Open to anyone at Dartmouth regardless of background, undergrad or grad.
Sign-ups close October 7. Time and place will be announced in the GroupMe.
Rough weekly topics (subject to change)
Week 0Introduction to machine learning (optional, self-paced)
Week 1Where AI is headed
Week 2Fundamentals of misalignment
Week 3Why would an AI want to do bad things?
Week 4Threat models
Week 5AI control and evaluations
Week 6Interpretability, careers, and getting involved