Governments Setting Limits on AI

The Biden Administration's actions came on the heels of the European Union, which last June passed the landmark Artificial Intelligence Act, moving a step closer to formally adopting the first-of-its-kind set of comprehensive rules around regulating AI. The AI Act, which was expected to be adopted early this year, sets four classifications for AI risk, ranging from minimal to unacceptable. Technology classified as an unacceptable risk, for example, would include systems that judge people based on a behavior known as social scoring, along with predictive policing tools, and would be banned. There also will be an EU AI board to oversee the implementation and uniform application of the regulations, which will build on existing GDPR and Intellectual Property legislation. The AI Act "is the first comprehensive regulation addressing the risks of artificial intelligence through a set of obligations and requirements that intend to safeguard the health, safety and fundamental rights of EU citizens and beyond, and is expected to have an outsized impact on AI governance worldwide," wrote Mia Hoffmann, a research fellow at the Center for Security and Emerging Technology (CSET) at Georgetown University.