Should AI labs be allowed self-police their models? Absolutely not, Microsoft’s Brad...
Should AI labs be allowed self-police their models? Absolutely not, Microsoft’s Brad Smith says
Topic
3 stories · No followers yet
Stories by day over the last 30 days.
Should AI labs be allowed self-police their models? Absolutely not, Microsoft’s Brad Smith says
Brad Smith calls for emergency brakes, mandatory incident reporting, independent safety evaluations amid growing debate over AI regulation
Brad Smith calls for global rules on AI, and says the UK is in the "best position" to lead the charge.