Jim the AI Whisperer

Jim coaches professionals and organisations on how to get the most out of generative AI tools without getting swept away by marketing myths, and is a provocative leading voice on the promises and perils of generative AI — especially the bits companies would rather you didn’t look too closely at. 

As a red-team prompt engineer (i.e. simulating real-world adversarial behavior to uncover vulnerabilities) and system prompt sleuth, Jim has spent the past five years poking holes in AI safety claims by doing what most users aren’t supposed to: asking the right questions the wrong way.

All of this work is grounded in a single question: If these models are safe, why do they need to hide so much? Jim brings a voice of caution wrapped in wit, advocacy grounded in research, and questions no PR firm wants to hear.