AI less regulated than milk, Microsoft president says in warning over safety
Brad Smith called for stronger AI safeguards, including liability for victims and ways to pause systems that behave dangerously.
TL;DR
- Brad Smith called for stronger AI safety duties and emergency stop mechanisms.
- UK oversight plans were under discussion, not enacted rules.
- The milk comparison is Smith’s argument, not a measured regulatory study.
Microsoft president Brad Smith told the BBC that ordinary products such as milk and washing machines face safety rules that AI does not. He argued for rules that include an obligation to victims when rogue AI agents cause harm and for emergency mechanisms that can pause or slow systems. [1] [2]
The Guardian reports that Smith described AI as both a tool and a potential weapon and called for urgent regulation. Microsoft CEO Satya Nadella separately advocated systems whose behavior and limits can be observed and tested, according to the same coverage. [1] [2]
The comments arrive amid calls from safety researchers and governments for more oversight of advanced systems. The report also notes that the UK government was preparing to call on frontier model developers to strengthen safeguards and was considering legislative oversight; those are proposals and planned remarks, not enacted rules. [1] [2]
The unresolved issue is what specific obligations, testing standards, and enforcement mechanisms policymakers will adopt. Smith’s comparison is an argument for regulation, not a measured comparison of regulatory codes across product sectors; the discussion does not establish a settled legal framework. [1] [2]
Why it matters
A leading AI company executive publicly supports safety regulation, but the content and status of any new legal requirements remain open.
Editor's note
The headline analogy is attributed to Smith; regulatory proposals are not described as enacted law.