AI Alignment

Ethics and Safety

The challenge of making AI systems reliably pursue what humans actually intend - not just what we literally asked for.

Alignment asks: how do we make sure powerful AI systems do what we mean, respect human values, and stay safe even in situations their designers never anticipated?Even today it is practical work - training assistants to refuse harmful requests, be honest about uncertainty, and avoid manipulation. As systems become more autonomous, alignment becomes one of the central problems of the field.