When a model learns your secrets, it never forgets.

In this episode, Nick sits down with Tom Gillis from Cisco to explore one of the most pressing challenges of the AI era: data sovereignty. As AI models absorb sensitive information and operational data at unprecedented scale, organizations are facing a new reality: once that knowledge is embedded, it can’t simply be erased.

Tom unpacks how Cisco is reimagining the data center for this new world from GPU-driven architectures and co-packaged optics to federated analytics that bring computation to the data instead of the other way around. Together, they discuss the return of on-prem infrastructure, the risks of IP leakage in model training, and why protecting data sovereignty may be the defining trust challenge of enterprise AI.

New episodes of the Built for Trust Podcast come out every Thursday. Subscribe to get notified. See you next week!

In this episode, Nick sits down with David Reilly from World Wide Technology to explore how trust and structure must guide every AI transformation. Riley shares the five-question framework he used to evaluate technology decisions, from cost and reliability to risk and talent, and why it’s more critical than ever as enterprises race to adopt AI. Together, they discuss how CIOs can balance innovation with stability, build confidence with business partners, and lead teams through change without losing trust along the way.

New episodes of the Built for Trust Podcast come out every Thursday. Subscribe to get notified. See you next week!

Hyperscalers have redefined what modern networking looks like, and now those same principles are reshaping how enterprises build for the AI era.

In this episode, Nick sits down with Mark Austin from Hedgehog to explore how open networking, SONiC, and cloud native automation are converging to make hyperscaler level networking accessible to organizations of any size.

Mark shares Hedgehog’s origin story, why SONiC has finally matured for enterprise scale, and how cloud style UX and zero touch lifecycle management remove the historical barriers to open networking. They explore the rise of AI Networking as a new category, the complexity of GPU fabrics, and how automation and performance tuning can even outperform some NVIDIA reference benchmarks.

From multi-vendor freedom to real world AI cloud deployments operated by a single DevOps engineer, this conversation reveals the technologies and design principles that are powering the next generation of private AI infrastructure.

If you are building for AI, rethinking your network strategy, or interested in the future of open networking, this episode shows what it really takes to network like a hyperscaler.

New episodes of the Built for Trust Podcast come out every Thursday. Subscribe to get notified. See you next week!