Apply to participate in a research sprint focused on AI alignment and AI governance. This event is an opportunity for those interested in exploring the realms of AI safety and governance, contributing to a vital area of research with significant potential impact. Our event aims to, amongst other things, contribute to the broader discussion on AI's future.

This talk will introduce the core challenges in aligning advanced AI to human values and preventing us losing control of these systems.

Marius Hobbhahn is the CEO of Apollo Research, a recently founded AI Safety organisation that is working on understanding AI systems well enough to prevent deceptive behaviour.

Patrick Levermore, currently working as a Risk Analyst at the UK government’s office for AI while on secondment from the Institute for AI Policy and Strategy. His talk will cover the risks related to development and deployment of frontier AI systems and the role policy and governance can play in reducing these risks.

With AI entering the mainstream and the largest and most capable models still being black box, we want to figure out what is going on inside them! With Alignment Jams and the great researcher Neel Nanda, we get a chance to create impactful and real research on interpreting neural networks.