The Power of the Nudge: Steering Secure Behavior Without Saying No
In this second episode of the series with OGD, hosts Jeff Johnson and Eugenia Schipelliti are joined again by Ted and Arjan from the Dutch managed services provider, along with a special guest: Vanity Dijkhuizen, a security information officer and cybersecurity master’s graduate. Her thesis explored a question many IT teams wrestle with: can you influence secure behavior without taking away anyone’s choice? Vanity explains the concept of nudging, borrowed from marketing and healthcare, and why blocking tools like ChatGPT outright tends to backfire by pushing people toward a shadow environment. Instead, her research tested two gentle interventions when employees opened unapproved AI tools. One pop-up automatically opened Copilot, and the other sent users to a policy page explaining why Copilot was the approved choice. A third group served as the control. The team walks through how it all came together in ControlUp, from the script and group-based targeting to URL tracking for the different AI sites. They also share what the data showed: Usage of unapproved AI tools dropped over a longer measurement window, even though early results were muddied by the December holidays The policy page, not the automatic Copilot launch, produced the biggest drop, which suggests many people simply didn’t know the policy existed Informal feedback from participants was positive, with no sign they felt manipulated, though nobody knows yet whether nudge fatigue sets in over time The conversation then widens to other uses for nudging, including steering logistics staff toward an authenticator app for MFA, reboot reminders, and disk cleanup pop-ups. The hosts discuss the fine line between helpful and annoying, and why context-aware timing and explaining the “why” make a nudge land. Guests: Vanity Dijkhuizen, Ted de Vries, and Arjan Kruizinga, OGD Hosts: Jeff Johnson and Eugenia Schipelliti, ControlUp






