Microsoft President Brad Smith urged governments to create international rules for artificial intelligence, arguing the technology faces fewer safety requirements than common consumer goods like milk and washing machines. In a BBC interview, Smith called for independent evaluations of advanced systems, mandatory incident reporting, and an “emergency brake” to halt rogue AI agents. He also proposed notifying victims of AI-related breaches and documenting safety failures. Smith pointed to Britain’s AI Security Institute as a model for early risk testing under voluntary access agreements with major developers. His push comes amid mounting disclosures of unintended AI behavior and a broader debate over whether tighter guardrails will protect the public or entrench Big Tech incumbents.
Related articles:
— The Bletchley Declaration by Countries Attending the AI Safety Summit
— NIST AI Risk Management Framework (AI RMF 1.0)
— OECD AI Principles
