How should we regulate artificial intelligence?

Council verdict: The Council agreed on a risk-tiered regulatory framework: light-touch oversight for low-stakes applications, mandatory audits and transparency requirements for high-stakes uses like criminal justice a…