Overview:
A working session for managers, team leads, and project professionals who need to put real AI guardrails in place without waiting for a corporate policy that may never arrive. We cover the small number of governance decisions that actually matter: data classification and what can be fed to which tools, approved versus unapproved use cases, human-in-the-loop requirements, disclosure norms, and incident response. The session is tool-agnostic and focuses on the rules and habits - not the products.
Why you should Attend:
Someone on your team is about to paste sensitive data into a chatbot, ship code an agent wrote without reviewing it, or send a client a deliverable a model hallucinated. If your only AI governance is 'be careful,' you are one bad afternoon away from a data breach, a compliance finding, or a public embarrassment. The time to write the rules is before the incident, not after.
Areas Covered in the Session:
- Why working-manager governance differs from enterprise AI policy - and why you cannot wait for the latter
- The four governance questions every team should be able to answer in writing
- Data classification basics: what AI tools can see, what they cannot, and how to decide
- Approved use cases, conditional use cases, and prohibited use cases - building your own short list
- Human-in-the-loop: where it is required, where it is theatre, and how to tell the difference
- Disclosure and attribution: when to tell clients, regulators, and teammates that AI was used
- Incident response: what to do when an AI mistake reaches a customer or a regulator
- A one-page AI operating agreement your team can adopt this week
Who Will Benefit:
- Working Managers and Department Heads
- Project Managers and Program Managers
- Operations Managers and Functional Leads
- Compliance Officers and Risk Managers
- Information Security Managers without a dedicated AI policy
- Directors of PMO and Heads of Delivery
- HR Managers and L&D leaders setting team norms