We are releasing the course materials of the Iliad Intensive, a new month-long, full-time AI alignment course that runs in-person every second month. The course targets students with strong backgrounds in mathematics, physics, or theoretical computer science, and the materials reflect that: they include mathematical exercises with solutions, self-contained lecture notes on topics like singular learning theory and data attribution, and coding problems, at a depth that is unmatched for many of the topics we cover.
Around 20 contributors were involved in developing these materials for the April 2026 cohort of the Iliad Intensive. By sharing the materials, we hope to:
- create more common knowledge about what the Iliad Intensive is;
- invite feedback on the materials;
- and allow others to learn via independent study.
We are developing the materials further and plan to eventually release them on a website that will be continuously maintained. We will also add, remove, and modify modules going forward to improve and expand the course over time.
Structure
The Iliad Intensive is structured into clusters — loose collections of related topics — which decompose into modules taught within one day. Each module consists of learning outcomes, prerequisites, and the content itself, including a fast-track and pointers for how to learn more. Some modules include a teaching guide that explains how the content was taught during the April Intensive.
- 0 Prerequisites. What is useful to know before engaging with the materials: a background worldview (why AI matters, safety risks) and the technical prerequisites — deep learning, linear algebra, calculus, probability & statistics, information theory, and some theoretical computer science.
- Cluster A — Alignment. The problem of how to align AI systems, or collections of such systems, with a vision for how they should behave.
- Cluster B — Learning. The principles and practice of (deep) learning — what we can say about learning machines in general, what's mysterious about deep nets specifically, and what frameworks (singular learning theory, training dynamics, data attribution) we have for understanding them.
- Cluster C — Abstractions, Representations, and Interpretability. The internal representations and mechanisms of cognitive systems: mechanistic interpretability, computational mechanics, abstractions and latents.
- Cluster D — Agency. Long-term goal-directed behaviour in unbounded environments — reinforcement learning, idealised agency, agent foundations, world models.
- Cluster E — Safety Guarantees and their Limits. Debate as an approach to direct alignment, and several limits to ensuring the transparency and safety of AI systems in general (steganography, backdoors, worst-case interpretability).
Feedback
Please provide feedback on the materials in the comments of the LessWrong post, or as an email to feedback@iliad.ac.