Series 10 of 16
AI Safety and Alignment
What the guardrails are, how they are built, and what they are for.
Choose how to learn
Episode 1 · comingWhat is alignment?Why does it refuse things?
Episode 2 · comingGuardrailsWhere do the rules actually live?
Episode 3 · comingJailbreaksWhy can the rules be talked around?
Episode 4 · comingThe risk argumentWho is worried about what, exactly?
Next seriesPrompting and Retrieval