Episode 23: Beyond Compliance: Rethinking Data Protection in an AI-Driven World

"It's not 'don't do,' it's 'how to.'"

James Robson, Founder of Datum Longevitus and former Data Protection Officer for the Labour Party reveals how organisations can use data protection to speed up AI rollouts rather than slow them down.

In this Data & AI Podcast conversation with Waseem Ali, they talk about: 

• Why DPOs need to be involved at the start of projects, and how to bring the wider business with you
• What organisations get wrong when rolling out Copilot and other LLM tools
• The ethics of automated decision-making and what transparency actually requires
• Why "human in the loop" still matters in AI-driven decision-making
• How poor AI configuration creates unnecessary data and IP risk

Waseem and James also explore real examples shaping the future of AI governance, including the Labour Party's data environment, the Housing Helper homelessness chatbot used by local councils, and the facial recognition platform PIM Eyes.

The episode closes with James's view on what's next: trusted data intermediaries that let organisations share data compliantly, and could reshape the AI tools built on top of it.

Plenty to take away for anyone working out where AI compliance ends and AI innovation begins. Subscribe to the Data & AI Podcast for more conversations on data, AI, and the future of work. If you found this useful, share it with someone navigating their own AI compliance work, or visit harnham.com to find out how we support data and AI hiring across the UK and US.