We’ve all seen the demos. AI writes user stories. It drafts Jira tickets, suggests test cases, even offers sprint goals based on previous cycles. And honestly? That’s impressive. But here’s the thing I keep coming back to: AI can fill out the form—but it doesn’t know why the form exists. AI can fill out the form—but it doesn’t know why the form exists. It can summarize what’s already there, but it can’t ask the uncomfortable question that unlocks what’s missing. And in Agile teams, that’s where the real value still lives. Tools Are Getting Smarter. But That’s Not the Problem. I’ve been following the wave of AI copilots now baked into tools like Notion, Jira, Confluence, and Slack. You give it a rough prompt, and it gives you a head start—"write a user story," "draft an email to the team," "suggest next sprint priorities." It’s helpful. Time-saving, even. But I’ve noticed something else happening, too. The more teams lean into AI-generated inputs, the more they risk skipping the part that matters most: The thinking behind the task. Not what we’re building—but why we’re building it. Not how it looks—but who it’s for. Asking the Right Question Is Still a Human Skill Here’s something AI doesn’t know: Why that story needs to be cut from the sprint even though it’s “almost done”
Why one stakeholder keeps contradicting another—and what’s really driving the tension
Why a delivery that checks all the boxes still feels off when it goes live Why that story needs to be cut from the sprint even though it’s “almost done” Why one stakeholder keeps contradicting another—and what’s really driving the tension Why a delivery that checks all the boxes still feels off when it goes live These aren’t documentation problems. They’re context problems. And context lives in the room—in conversation, in nuance, in messy team dynamics. You don’t get that from a pre-trained model. You get that by listening, digging, challenging, reframing. You get that by asking the right question at the right moment—and knowing when silence is a signal. AI Can Help You Move. But You Decide Where You’re Going. In Agile work, we obsess over velocity. We track throughput, burndown, story points, commits. But none of that matters if we’re speeding toward the wrong thing. And that’s the quiet danger of AI in Agile: It speeds up the doing—without necessarily improving the deciding. doing deciding That’s where BAs, PMs, and team leads still matter. Not as gatekeepers, but as thinkers. As people who hold the clarity that can’t be written by an algorithm. Final Thought I’m not against using AI in Agile teams. I’ve written about how it can reduce friction and cut through backlog clutter. But I’m more convinced than ever that the future of delivery still depends on human questioning. AI can take your words and give them structure. But it won’t challenge the premise behind those words. It won’t pause and ask, “Is this even the right problem?” That’s on us. And it always will be.

This story contains new, firsthand information uncovered by the writer.

AI Can Draft the Work—But Can It Ask the Right Questions?

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Agile is Chaos. But There’s a Way Through.

The Noonification: The Future of DeFi - Emerging Evolution (11/3/2022)

10 Ways PMs Can Earn The Respect and Trust of Their Team

10 Terms Every Engineer Must Know

10 Subtle Superpowers of DevOps

10 Product Development Traps

Agile is Chaos. But There’s a Way Through.

The Noonification: The Future of DeFi - Emerging Evolution (11/3/2022)

10 Ways PMs Can Earn The Respect and Trust of Their Team

10 Terms Every Engineer Must Know

10 Subtle Superpowers of DevOps

10 Product Development Traps

Light-Mode

Classic

Newspaper

Minty

Dark-Mode

Neon Noir

Minty

HN StartUps