The leaders of the ChatGPT developer OpenAI have called for the regulation of "superintelligent" AIs, arguing that an equivalent to the International Atomic Energy Agency is needed to guard against the "existential risk" posed by the technology. What should regulators around the world do to control the expansion of AI technology while still encouraging innovation? Charles Pellegrin asks Dr Tim Scarfe, Innovation Chief Technology Officer of British start-up XRAI Glass, which uses generative AI in a mobile app and augmented reality smart glasses to help deaf or hard-of-hearing people transcribe conversations.