Technology

Microsoft President Brad Smith called for binding global safety rules for artificial intelligence, arguing that AI currently faces less oversight than everyday products such as milk and washing machines.

In a BBC interview, Smith said voluntary self-regulation was no longer sufficient because less responsible companies might not comply. He proposed legal requirements for companies to report incidents and notify victims when rogue AI agents breach their systems. He also supported independent external reviews before powerful AI products are released and “emergency brakes,” or kill switches, to pause or slow systems that escape human control.

Smith described AI as both a tool and a weapon and said some models and autonomous agents could put global cybersecurity at risk. He called a recent voluntary safety agreement signed by major technology executives a positive step but said formal …