David Dalrymple on Safeguarded, Transformative AI

David "davidad" Dalrymple joins the podcast to explore Safeguarded AI — an approach to ensuring the safety of highly advanced AI systems. We discuss the structure and layers of Safeguarded AI, how to formalize more aspects of the world, and how to build safety into computer hardware.  

You can learn more about David's work at ARIA here:   

https://www.aria.org.uk/opportunity-spaces/mathematics-for-safe-ai/safeguarded-ai/   

Timestamps:  

00:00 What is Safeguarded AI?  

16:28 Implementing Safeguarded AI 

22:58 Can we trust Safeguarded AIs?  

31:00 Formalizing more of the world  

37:34 The performance cost of verified AI  

47:58 Changing attitudes towards AI  

52:39 Flexible‬‭ Hardware-Enabled‬‭ Guarantees 

01:24:15 Mind uploading  

01:36:14 Lessons from David's early life