What You'll Learn

Explore ethical, social, and technical challenges in AI. Aim for alignment with human values and societal norms.

- Develop a benchmark to track and improve AI model and prompt performance over time. - Use moderation models to evaluate and score harmful AI outputs. - Train and prompt engineer AI models towards or away from specific values. - Create a values-evaluation model through self-consistency. - Understand and discuss tokens, values, ethics, reward misspecification, and scalable oversight. - Apply techniques to reduce AI hallucination, ensure AI confidentiality, and detect sleeper agents.

Course Schedule

4 Weeks
12.0 Classroom Hours

Week 1 2 sessions

Ethics and AI Alignment

Tuesday, October 29 at 5:00 PM - 6:30 PM PDT

Engage in discussions on AI ethics, model benchmarking, and alignment strategies.

1.5h

Benchmarking and Moderation Models

Thursday, October 31 at 4:00 PM - 5:30 PM PDT

Participate in workshops to apply AI alignment strategies in practical scenarios.

1.5h

Week 2 2 sessions

Sector Impacts of AI Alignment

Tuesday, November 05 at 5:00 PM - 6:30 PM PST

Discuss the implications of AI alignment in various sectors and the importance of ethical considerations.

1.5h

Practical AI Alignment Strategies

Thursday, November 07 at 5:00 PM - 6:30 PM PST

Hands-on workshop on developing benchmarks and using moderation models.

1.5h

Week 3 2 sessions

Challenges in Value Alignment

Tuesday, November 12 at 5:00 PM - 6:30 PM PST

Explore the challenges and solutions in prompt engineering for value alignment.

1.5h

Values-Evaluation Modeling

Thursday, November 14 at 5:00 PM - 6:30 PM PST

Workshop on creating a values-evaluation model through self-consistency.

1.5h

Week 4 2 sessions

Insights and Directions in AI

Tuesday, November 19 at 5:00 PM - 6:30 PM PST

Final discussions on course learnings, insights, and future directions in AI alignment.

1.5h

AI Alignment: A Comprehensive Overview

Thursday, November 21 at 5:00 PM - 6:30 PM PST

Conclude the course with a comprehensive discussion on AI alignment, summarizing key learnings and future prospects.

1.5h

All times shown in Pacific Time (PT)

Who Is This Course For

Experienced developers, managers, and executives

Join the Course

Pay what you can - accessible learning for all

We use a suggested price to prevent abuse, but if you need a larger discount, email us at liz@themultiverse.school. We offer 100% scholarships - no one is turned away for lack of funds.

$ 200

Suggested price • Pay what you can

Join Now

Secure payment via Stripe

$ 240

Suggested price • Pay what you can

Join Now

Secure payment via Stripe

Want to schedule this course for your team?

Contact us: liz@themultiverse.school