As we repeat the course, we’ll release more lecture resources.

You can see the most up-to-date version of the resources on this google drive.

We’ll closely follow the structure and content from the original version of the course, and bring each lecture up to date with the (fast) progress in the field over the last five months. Please check out the recordings - then join us for the latest perspectives, and the chance to ask questions live!

We are also planning to run examples classes in person in Cambridge to accompany the course: we will release problems sheets a week before each examples class and then use the examples class to facilitate small-group discussion of the problems. Provisionally these will be Saturdays at 2pm every two weeks from Sat 25th October till Sat 6th December. Sign up to the mailing list, or check this google doc for the latest updates.

The lectures and example classes will be in person only, but will be accompanied by additional lecture resources that we will release publicly online. These resources will also complement the recordings from the original version of the course.

You can see the most up-to-date version of the resources on this google drive.

What you’ll learn

This 16 lecture series will explain how language-model systems are built in order to understand and predict their behaviour. Frontier language models are now being used as the foundation for agentic systems, which can carry out tasks that require extended reasoning and long-horizon planning. We investigate the potential safety and security risks associated with such systems, and present current research directions that aim to mitigate them.

The series is designed to be accessible for a broad audience across academia and industry, requiring knowledge from an introductory course in machine learning or statistics (e.g. backpropagation). We emphasise conceptual understanding of such systems, but will discuss technical details where necessary.

We hope that the course will empower researchers to make better use of language model systems and inform deployment across academia and industry. We also hope to stimulate engagement with the serious risks associated with intelligent systems, and encourage further work to address them.

Online course

The video recordings from the Summer 2025 edition are available below.