This form is the permanent interest form for the AI Alignment McGill (AIAM)-run AGI Safety Fundamentals reading group. We expect this application to take 5 minutes.
This programme focuses on the alignment and control problems surrounding future artificial general intelligence. Topics include AI interpretability and explainability, reward learning, misalignment in large language models, careers in AI safety, and more.
If you're interested in learning more about AI safety to get a sense of this program, we recommend starting with the introductory materials found in our resources: https://agisf.com/resources.
The program is based on a curriculum by Richard Ngo, a researcher at OpenAI.
The reading group will run for seven weeks, date TBD. This program involves weekly in-person one-hour discussions in small groups of 3-6 participants and one discussion facilitator, as well as 1–2 hours of readings and exercises before discussion meetings. You can view a list of topics in this program at https://course.aisafetyfundamentals.com/alignment, which we will use as our syllabus.
Discussion group meeting times will be scheduled at a time when you are available.
If you have any questions, feel free to email us at alignmentmcgill@gmail.com.
Join the AI Alignment McGill discord server: https://discord.gg/8F5NKntrDt