How are you using AI in your agile work?

Michael Brinn
We've been exploring AI not just as a tool, but as a collaborative team member — one that supports cognition, learns with us, and even helps shape better team dynamics over time.

- Retrospectives + Emotional Tone Mapping
We’ve used AI to analyze the emotional tone of team feedback (anonymized and voluntary), identifying early signs of burnout, misalignment, or interpersonal tension.
This has become a powerful enabler of psychological safety, in line with Amy Edmondson’s work — fostering healthier retros and regenerative team rhythms.

- Backlog Refinement + Thematic Clustering
AI scans user feedback, historical stories, and technical debt to reveal emergent themes and weak signals we might miss.
It acts almost like a cognitive twin — not faster than humans, but broader in pattern recognition.
This mirrors the “externalize–combine” phases of SECI 2.0, accelerating actionable insight.

- Sprint Planning with Scenario Simulation
By blending historical velocity with dynamic constraints, AI helps us simulate realistic sprint outcomes. It’s like having a Cynefin-informed co-pilot — offering clarity when navigating complexity and reducing planning friction.

Quick Win
One of our best experiments: using AI to generate acceptance criteria based on historical stories and stakeholder patterns.
Result: 30% drop in story rework and better alignment between POs and developers — with minimal setup and zero disruption.

Key Insight
When AI is embraced as a learning partner, not just automation, it deepens reflection, foresight, and trust.
We now design processes where humans and AI co-evolve, building what we call augmented agility — a mindset that blends speed with sensemaking.

This shift aligns with regenerative leadership: we don’t use AI to replace judgment, but to amplify our ability to learn, adapt, and collaborate across evolving contexts.

Would love to hear how others are integrating AI into the team, not just into the tools — especially in ways that build capacity, connection, and clarity.

AI is a board term. Organizations are using AI supporting agile approach from more than 40 years ago. Just to remember, agile is not about to use a method, is not about creating software products, things like that. Agile was defined in 1990 to be used in manufacturing. Now, with the publishing of generative AI in 2017 and associated tools like ChatGPT you can use "the universal knowledge and practice" to create artifacts that match for the agile based method you are using to create software products. Scrum for example. The only thing you need to do is to use the right framework: https://www.linkedin.com/posts/khizer-abba...607471104-Jv76/ asking ChatGPT (for example, or others) to create the artifacts and do not forget to use the same frameworks to do things like consistency and trust checking. With all that said, remember the impacts to do not use Responsible AI frameworks and mainly always remember: human in the loop is the key success factor when use AI.

Luis,
I really appreciate how you’ve connected it to concepts like psychological safety. It reframes AI not as a bolt-on tool, but as part of the team’s learning system.

A couple of things especially resonate with me:

Tone mapping in retros. I’ve seen teams struggle to surface “what’s really being felt” until it’s too late. The idea of AI amplifying emotional signals without judgment seems like a game changer for sustaining healthier rhythms.

Scenario simulation in planning: This feels like where AI can add real foresight. Instead of arguing over estimates, teams can explore modeled outcomes and then choose based on context, values, and risk appetite.

Thanks so much for sharing this!

But I think your point about responsible AI and human-in-the-loop is the most important one: without guardrails for consistency, trust, and ethical use, we risk undermining the very values agile is meant to uphold.

Thank you so much for sharing these ideas and frameworks.

I’ve found the biggest wins when using AI for backlog refinement and customer feedback analysis. It helps me spot patterns, cluster themes, and surface dependencies faster, which gives the team more time to focus on real discussions instead of prep work. Also used AI to draft retrospective prompts or summarize complex stakeholder feedback before a planning session. Nothing replaces the team’s judgment, but AI speeds up the groundwork so we start from a clearer baseline.