We describe the basic components of Responsible Scaling Policies (RSPs) as well as why we find them promising for reducing catastrophic risks from AI.