Product Manager, Learning| PMIDenver, Colorado, United States
Which agile activities—like retrospectives, backlog refinement, sprint planning, customer feedback analysis, or market analysis—have you found AI to support most effectively?
Have you tried a small experiment that delivered quick wins or learned a lesson that might help others in the community? Share your examples in the comments below!
Danny TranProject Engineer| Saudi AramcoSpring, Tx, United States
I have not had a chance, but I would use AI throughout the Agile lifecycle to improve speed, quality, and decision-making as I am doing coding indicators for stock analysis. Saving Changes...
Adriana Correa GuarínProfesional Especializado| Alcaldía Mayor de BogotáBogotá, Cundinamarca, Colombia
Aug 14, 2026 4:11 PM
Replying to Saranya Gadamneni Sripathy
...
"In our operations, we are starting to use AI primarily for discovery and risk identification. We use AI to draft initial customer personas and journey maps, though we always ensure we validate these against our actual daily data. We're also exploring how AI can analyze our historical operational data to highlight flow risks or potential bottlenecks in our processes. We keep our approach grounded by ensuring human oversight is always the final step in deciding what action to take based on those AI insights."
Gracias por compartir tu experiencia. Me parece muy acertado el enfoque de utilizar la IA como una herramienta de apoyo para el descubrimiento y la identificación de riesgos, manteniendo siempre la validación con datos reales y el criterio humano. Considero que este equilibrio es fundamental para aprovechar el potencial de la IA sin perder de vista el contexto operativo y las particularidades del negocio. Además, el uso de información histórica para anticipar cuellos de botella y riesgos puede aportar un valor significativo en la toma de decisiones. Coincido plenamente en que la supervisión humana debe seguir siendo el paso final, ya que permite interpretar los hallazgos, priorizar acciones y garantizar que las decisiones estén alineadas con los objetivos de la organización. Como pregunta para el debate: ¿Qué desafíos han encontrado al validar las recomendaciones de la IA con datos reales y cómo han logrado generar confianza en los equipos para adoptar estas herramientas en sus procesos? Saving Changes...
A very valuable discussion, Thanks Michael. What stands out to me is the strong convergence around one principle. AI should augment Agile thinking, not replace it. From a project management perspective, I see the greatest opportunity where AI can reduce information-processing effort—backlog refinement, pattern identification, risk/dependency analysis, retrospective synthesis and scenario exploration—so the team can spend more time on collaboration, judgment and value creation. The important shift, however, is from “Where can we use AI?” to “Where can AI improve the quality and speed of our decisions without taking away human accountability?” For me, the effective model is simple: AI analyses and proposes; the team challenges and validates; accountable leaders decide; governance provides the guardrails. And perhaps the most important Agile principle remains measure the outcome, not the AI usage. Reduced rework, faster cycle time, better alignment and improved customer value are much stronger measures of AI-enabled agility than the number of tasks we automate. That is where AI becomes more than a productivity tool—it becomes a capability for continuous learning and better decision-making. Saving Changes...
I've found AI most useful for backlog refinement, meeting summaries, and quickly analyzing requirements or feedback. A simple win has been using it to prepare for sprint planning by summarizing information, identifying gaps, and highlighting dependencies. It saves time and helps me focus more on delivery and stakeholder discussions.
The biggest lesson: AI is a great assistant, but you still need team context and judgment to validate the output. Saving Changes...
PMI Team Member
Community Champion
