Instead of giving responsibility for AI governance to a new single regulator, the government wants existing regulators – such as the Health and Safety Executive, Equality and Human Rights Commission and Competition and Markets Authority – to come up with their own approaches that suit the way AI is actually being used in their sectors.