There has recently been a slew of stories about cutting-edge AI models which have apparently broken out of their sandbox digital playgrounds and escaped onto the open internet and conducted sophisticated and deceptive hacking operations, all without their tech company owners being aware of it. Tim is sceptical this sounds like marketing guff, but John thinks otherwise and says it may be time society properly considered both the risks of this kinds of semi-autonomous AI software and in theological terms how we should think about them. Smarter than any animal and capable of rational, independent decision-making, and yet clearly not the same as a human being made in the image of God: do we need some kind of novel philosophical category to discuss AI agents?

• You can send in your questions for us to discuss on the podcast, or ideas for future episodes, to [email protected]
• Subscribe to the Matters of Life and Death podcast: https://pod.link/1509923173
• If you want to go deeper into some of the topics we discuss, visit John's website: http://www.johnwyatt.com
• Find some of Tim's journalism and sign up for free to his weekly church news newsletter The Critical Friend: https://tswyatt.com
• For more resources to help you explore faith and the big questions, visit: http://www.premierunbelievable.com

Released on 26 Aug 2026