AI in Agile: Fad, Silver Bullet, or Evolution?

Agile has always promised adaptability, customer value, and continuous improvement, but its history is riddled with executive-driven silver bullets. From the overuse of the Spotify model to the rigid application of SAFe, we’ve seen organizations latch onto the “next big thing,” hoping to fix deep systemic issues with superficial solutions. It didn’t help that agile purveyors wrote books like Scrum: The Art of Doing Twice the Work in Half the Time, which could lead to unrealistic expectations. 

Now, AI, particularly generative AI, is generating a similar buzz. Is this just another Agile fad, or could it signal a true evolution in the way we work?

Agile's History of Fads

Before we examine AI’s role, it’s worth reflecting on the agile anti-patterns that have emerged over time, anti-patterns of hype, shallow adoption, and inevitable disillusionment:

• Spotify Model Copycats
  Inspired by conference talks and case studies, companies copied the names (tribes, squads, chapters, guilds) without copying the intent, context, or supporting culture. The model was not meant to be a universal framework, but was a snapshot of something Spotify tried at a particular time. When applied blindly, it often led to confusion and fragmentation rather than agility. Agile requires learning about and adapting to each unique environment.

• SAFe Misuse
  SAFe was designed to help large enterprises apply Agile at scale, but it’s often used to justify top-down control and heavyweight governance. Many organizations cherry-pick the most hierarchical elements, mistaking structure for agility, and lose sight of lean principles. The result can be bureaucracy and micromanagement under an agile banner.

• Zombie Scrum
  Teams follow the rituals: daily stand-ups, sprints, reviews, but miss the heart of Scrum: transparency, inspection, and adaptation. Without a culture of empiricism and ownership, Scrum becomes mechanical. It’s not unusual to see teams go through the motions while continuing to deliver low value at high cost. To quote Simon Sinek, Start With Why.
  

• Jira-as-Agile
  Many teams believe that having stories in Jira or using a Kanban board makes them agile. But Jira is a tracking tool, not a mindset. When teams focus more on being ticket-takers than on delighting their customers, agility is reduced to project management instead of product development.

• Paper Agility
  As capital-A Agile grew in popularity, certifications became a booming industry. But too often, teams and leaders equate certificates with competence. The result is paper agility: titles without transformation, and theory without practice. Welcome to the Agile Industrial Complex.

• Velocity Worship
  Story points were intended as a relative measure of effort, not a productivity KPI. But when teams are judged by how much and how fast they "burn down," they begin inflating points or gaming the system. Velocity becomes a vanity metric, disconnected from value delivery or sustainable pace. The 10th agile principle, “Simplicity – the art of maximizing the amount of work not done – is essential,” is all but forgotten.

• "Agile Coach as Savior"
  Organizations sometimes expect a coach to "fix the teams" while leadership continues operating with old habits and mindsets. The coach becomes a symbolic hire with little systemic support. Without broader change, coaching becomes isolated and ineffective.

• Scaling Before Maturity
  Enterprises often rush to scale agile without mastering it at the team level. They adopt LeSS, SAFe, or Nexus prematurely, creating layers of coordination before solving basic delivery problems. The complexity of scale magnifies dysfunction instead of solving it.

• Kanban Boards Without Flow Thinking
  Many teams adopt Kanban visually but ignore WIP limits, cycle times, or pull policies. The board becomes a static display of tasks rather than a tool for continuous improvement. Without flow thinking, Kanban loses its power to expose bottlenecks and improve throughput.

• “Agile Means No Planning”
  Some interpret agile as the absence of planning, leading to chaos instead of responsiveness. Good Agile involves frequent, adaptive planning, not neglecting it altogether. Skipping planning leads to unclear priorities, rework, and team frustration.
  

Each of these trends followed similar arc:

          Excitement → Shallow Adoption → Disappointment → Cynicism

Organizations hoped for transformation, but got templates. They sought simplicity, but ended up with stagnation. And now, AI risks following the same path, unless we learn from the past.

How AI Echoes Those Fads

Executive Hype Without Understanding
Just like with Spotify and SAFe, executives are jumping on AI because it’s trending. “Let’s use AI to write our user stories!” is today’s equivalent of “Let’s implement tribes!” often without changing the underlying behaviors that cause waste and stagnation. Without understanding how AI fits into actual team workflows, it becomes window dressing.

Misuse Leads to Disillusionment
We’ve already seen teams use ChatGPT or other LLMs to generate product backlog items, acceptance criteria, sprint plans, test cases, or even write code, all without critical thinking or validation. Much like past fads, AI gets blamed when poor implementation leads to poor results.

Tool Obsession Again
Just like Jira and Scrum boards were mistaken for agility, AI tools are at risk of being treated as ends in themselves. Teams automate away critical conversations or decisions in the name of "efficiency."  AI templates are used to run Daily Scrum and perform Retrospectives. This creates noise instead of clarity and destroys the first core value of the Manifesto, “Individuals and interactions over processes and tools.” 

AI is only as useful as the intent and collaboration behind it.

Where AI Could Be Different

AI as an Amplifier of Agility
Rather than impose structure, AI supports agility by enhancing adaptability. Teams can use AI to rapidly explore options, generate user story drafts, simulate impact, or analyze customer feedback. It doesn’t replace team decisions, but it gives teams more leverage in how they make them. Unlike frameworks that constrain teams, AI can expand their capacity to deliver value faster and with better information.

Prompting as a New Agile Literacy
Effective AI use isn’t about memorizing commands, it’s about communicating intent clearly, iterating toward better outcomes, and inspecting results. That’s the essence of agility. Developing strong prompting skills teaches teams to be concise, reflective, and iterative, exactly the habits we foster in story writing, test design, or retrospectives. Prompting is a team skill, not an individual hack.

Reflection and Learning at Scale
AI can surface insights across sprints, highlight themes from retrospectives, or suggest patterns in defect trends. This enables a kind of organizational sense-making that’s hard to do manually. Teams and leaders can use AI to see around corners, identifying issues earlier and adapting more quickly. Unlike static frameworks, AI offers living feedback loops.

Human + AI Collaboration Can Shift Mindsets
AI isn’t agile on its own; it becomes agile when it fosters Human creativity, improves transparency, and supports experimentation. Teams that work with AI collaboratively, rather than delegating blindly, can achieve higher learning velocity. This isn't about replacing team members; it's about augmenting their thinking and unblocking their flow. The key is that Humans have to have conversations and then incorporate AI into the conversation with them. There’s a dynamic in Human team conversations that occurs as part of story refinement. To effectively bring the AI tool into that conversation, use prompting and iterations, the same as the back-and-forth conversations that Human team members have when they work together on a problem.

Are AI Prompting Skills the Next Step to Agility?

Yes, but only if we approach them with the same discipline and mindset we bring to retrospectives, customer collaboration, and continuous improvement. Prompting isn't about clever hacks or shortcuts. It's about clarity of thought, precision of language, and intentional iteration. In that way, prompting is a microcosm of agile thinking: it forces us to get clear about what we want, to test assumptions, and to learn quickly from feedback. This is agility at its core.

Good prompting requires the same skills we develop when writing effective user stories or leading valuable sprint reviews. It demands that we distill complex needs into actionable inputs. It invites us to test variations, compare outputs, and adapt based on what we learn. This is not automation for the sake of speed, it’s collaboration with a machine in a way that mirrors our collaboration with teammates and stakeholders.

As teams learn to work with AI interactively, refining prompts together, evaluating the usefulness of results, and tuning based on context, they're building a new kind of shared literacy. This is where prompting becomes a team skill, not a one-off individual trick. When teams treat it as part of their workflow, it augments creativity, speeds up discovery, and sharpens focus. It becomes a muscle that strengthens agility, not a crutch that replaces thinking.

But if we treat prompting as magic, expecting perfect answers from vague inputs, or using AI to offload all the cognitive work, we fall into the same trap as past fads: we outsource responsibility rather than improve capability. Prompting skills will only support agility if teams maintain ownership of the outcomes, remain curious about the process, and treat every AI interaction as part of a broader feedback loop.

In this way, prompting isn’t just a tool; it’s a new frontier of agile practice.

The Bottom Line

AI won’t fix your product strategy, clarify your priorities, or create customer delight on its own. Agile won't fix them either, when considered a silver bullet. The point of agile isn't about transformation for transformation’s sake; it’s about changing how we build products, learn, and lead.

At its best, agility helps organizations work empirically: embracing transparency, shortening feedback loops, and making better decisions based on evidence, not just instinct. It encourages a shift from rigid planning to adaptive learning, from silos to shared ownership, and from gut feelings to insight-driven action.

Used wisely, AI can accelerate this shift. It can surface insights faster, expand creative possibilities, and support more informed decision-making. But it will only be effective in organizations that are ready to ask better questions, collaborate openly, and stay focused on what truly matters: delivering value.

In that sense, AI isn’t a silver bullet, it’s a new lever. And like any lever, its impact depends on where and how you apply it.

AI might not be just another fad; it could be the most agile tool we’ve seen yet.