Most retrospectives follow the same familiar pattern: people add notes to a board, group similar ideas, vote, and then rush through the most important conversation before time runs out.

What if AI could help the team have a better conversation, sooner?

In this session, we will explore a different approach to retrospectives. Instead of waiting for the meeting to uncover what matters, attendees will learn how AI can...

* Interview team members privately
* Identify patterns
* Help the facilitator focus the group on the issues worth discussing
* Suggest relevant "wonder" questions that help the team go deeper
* Surface possible actions the group may have missed

But giving AI a seat in the retrospective raises important questions. What should it know? What should remain private? When does it help the facilitator, and when does it get in the way? Can it help people speak more honestly, or does it make an already human practice feel less human?

You will leave with a practical framework for deciding where AI belongs in a retrospective, where it does not, and how Scrum Masters, agile coaches, and project managers can use it without giving up the judgment and trust that make retrospectives work.

(This talk belongs in the "AI Augmented Agile Delivery" track)