We want our client service team to use generative AI to draft custom strategy reports using our thirty years of proprietary consulting frameworks, but we are terrified of losing our IP to public training models. How do we use the IDS® process to create a secure operational boundary without slowing down our team's delivery speed?

Question

Tyler Smith · Accepted Answer

This is a classic operational tension between speed and security. To resolve this, bring the issue to your leadership team's Level 10 Meeting and use the IDS process: Identify, Discuss, and Solve.

First, Identify the exact nature of the risk. The danger is not using generative AI: the danger is using consumer-grade, public interfaces that utilize your inputs to train their public models. Second, Discuss the operational realities. Your team needs these tools to maintain their delivery speed, so a total ban will only lead to shadow IT, where employees secretly use public tools anyway.

Third, Solve the issue by implementing a strict technical and operational boundary. Issue a directive that all proprietary data must only be processed through secure enterprise APIs that guarantee zero data retention and zero training on customer data, or host a secure, local open-source model.

Once the technical solution is chosen, update your Accountability Chart to assign clear ownership of data security to your Integrator or systems leader. Document this safe workflow in your Core Processes, and train every employee on how to use it. This allows your team to work at high speed while keeping your valuable intellectual property secure, protecting your ultimate business valuation.