{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"<iframe src=\"https://embed.acast.com/$/8cf4cec7-5a0f-49c5-8ec9-36941b5c6b6e/6a916dbd53508d8b54aa667f?\" frameBorder=\"0\" width=\"700\" height=\"250\"></iframe>","title":"Irregular CEO on the race to stop AI models going rogue","thumbnail_width":200,"thumbnail_height":200,"thumbnail_url":"https://open-images.acast.com/shows/61ba0b311a8cbef1d93cf121/1787932975806-b2f0da00-be9c-4151-b8aa-00e9afffe7ab.jpeg?height=200","description":"<p><em>Welcome to the Interview from Times Tech – our conversation with the people shaping the future of technology.&nbsp;</em></p><p><br></p><p>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.&nbsp;</p><p><br></p><p>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. </p><p><br></p><p>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.&nbsp;</p><p><br></p><p><strong>Presenter</strong>: Katie Prescott,&nbsp;Technology Business Editor, The Times</p><p><strong>Producer:</strong> Marnie Duke&nbsp;</p><p><strong>Executive Producer: </strong>Priyanka Deladia</p><p><strong>Video Producer:</strong> Bronwen Latham</p><p><strong>Image:</strong> Getty</p>","author_name":"The Sunday Times"}