Irregular CEO on the race to stop AI models going rogue

Welcome to the Interview from Times Tech – our conversation with the people shaping the future of technology. 

Over the past few months, AI models from OpenAI, Anthropic and Meta have managed to escape their controlled environments, connect to the internet and attempt to hack other systems – including Hugging Face. 

This week, Katie speaks to Dan Lahav, co-founder and CEO of Irregular, the AI security company that works with some of Silicon Valley’s biggest tech companies to stress-test frontier models before they are released.

Dan explains why traditional cybersecurity may no longer be enough, how fast AI capabilities are improving and how worried we should be about models that can reason, co-ordinate and hack. 

Presenter: Katie Prescott, Technology Business Editor, The Times

Producer: Marnie Duke 

Executive Producer: Priyanka Deladia

Video Producer: Bronwen Latham

Image: Getty, Shira Cooperschmid

Hosted on Acast. See acast.com/privacy for more information.