The death of Agile teams: Why AI Is making teamwork obsolete

latest blog

Imagine it’s 2030 and an entrepreneur sits in a coffee shop with her laptop open. She sips her coffee while casually instructing her AI assistant to build a new product.

Within minutes, the AI generates a business strategy, drafts a marketing campaign, and spins up a working prototype. By lunchtime, she has launched an entire startup without a single employee. By the end of the week, she’s reaping the rewards of her idea and is dining at the Ritz.

This isn’t science fiction. It’s already happening.

OpenAI CEO Sam Altman recently stated that the first billion-dollar company run entirely by one person is inevitable.

For decades, we’ve been told that teamwork is essential for success. You may remember being told “there’s no 'I' in Team.” Corporate leadership programs teach us that successful companies are made up of great teams and collaborative teamwork. That innovation happens in groups, not alone.

But what if we got it all wrong? What if teams slow us down? What if AI can replace them?

Having worked in the field of ‘Agile project management’ for the past 20 years, I've witnessed first hand how small, cross-functional teams can drive game-changing results. But if AI can outperform teams in speed, efficiency, and innovation, where does that leave the principles of collaboration, agility, and human connection that we’ve spent decades refining?

Is this the end of the traditional Agile team?

One of the core Agile values emphasizes "individuals and interactions over processes and tools." However, large teams are no longer required for exponential value creation.

• Instagram had just 13 employees when Facebook acquired it for $1 billion.
• WhatsApp had only 55 employees when it was sold for $19 billion.
• Plenty of Fish, the dating website, was run by a single founder, making millions in profit with almost no staff.

These weren’t anomalies. They were signs of what’s coming.

Henrik Kniberg, a leading voice in the Agile community, recently wrote that cross-functional teams – the backbone of Agile – are no longer as essential as they once were. The reason? AI is rapidly replacing the need for diverse skill sets within teams.

"With generative AI, every person effectively has an AI colleague who is blazingly fast and knows every programming language, every framework, every design pattern, and possesses vastly more knowledge than any person," Kniberg wrote.

Instead of needing a team of developers, designers, and marketers, one person with AI can do everything. AI-powered businesses won’t eliminate collaboration entirely – but they have an opportunity to downsize teams.

More people, more problems

We assume more people = more productivity.

Wrong.

More people = more meetings, more complexity, more friction.

According to Brooks’ Law, adding people to a project makes it slower because communication overhead increases exponentially.

• A 5-person team has 10 lines of communication

• A 10-person team has 45 lines of communication

• A 15-person team has 105 lines of communication

The larger the team, the slower the decisions.

Jeff Bezos recognized this problem years ago. He famously created the “Two-Pizza Rule” – if a team can’t be fed with two pizzas, it’s too big.

Now, AI is making even two-pizza teams seem excessive.

Kniberg argues that Agile teams are shrinking. Instead of 8-10 people, AI-augmented teams could be just 1-2 people plus AI.

Why not one person? "Because it’s nice to have another human to talk to," he admits.

Is experience overrated?

For decades, we’ve hired for experience. More knowledge, more value. That was the belief.

But what if that belief is wrong?

Economist Ben Jones coined the burden of knowledge – the idea that as industries advance, innovation actually slows. Why? Because the more we learn, the harder it becomes to take risks, move fast, or see problems with fresh eyes. What once made experts invaluable now makes them cautious.

Traditional Agile thinking advocates we build cross-functional teams and distribute knowledge. No one person can know everything, so we split work across specialists. But now, AI bridges those gaps.

One of my first tasks as a Product Owner in the early 2000’s was to feed product teams with user stories. I was struggling to keep up because the team had little experience of writing them. So we hired four Business Analysts to help. Fast forward to 2025 and I can instruct AI to create them in an instant.

The role of today’s Product Owner is to focus on having the right conversations to better understand the real needs of your users and is less about crafting beautiful user stories.

AI will reshape teams, but it won't replace them

If we were to run a retrospective on what has worked well in Agile teams over the past two decades, I’m confident teamwork and collaboration would be popular themes. To abandon the concept of teamwork in Agile is like giving a Formula 1 driver a driverless car – we’re removing the very thing that gives us joy and a sense of purpose.

There’s no doubt that the rapid improvement in (especially generative) AI is forcing the Agile community to rethink how we work, but the real question isn’t whether Agile teams will survive, but whether Agile itself is agile enough to evolve. If it’s truly about adaptability, then it must embrace AI.

I still believe Agile teams matter regardless of the productivity gains AI will unlock. If in doubt, ask yourself what do you love most about where you work? I bet it’s not because you use the shiniest tools. We enjoy the work because of the people we work with. And the best places to work build deep and meaningful human connections. They create:

• Trust and psychological safety – Great teams share a sense of purpose. They’ve got each other’s backs. AI can’t replace that (at least, not yet).

• Creativity through debate – Innovation is about collectively solving a challenge, discussion, and building upon each other’s ideas.

• Human accountability – AI can generate endless ideas, but who decides which ones actually get built? Who makes the ethical calls? That’s still our responsibility.

• Meaning and belonging – The biggest risk of AI isn’t job loss, it’s loneliness. Work has never just been about output. It’s about connection. Strong work relationships improve happiness, retention, and mental health.

The problem isn’t teams. The problem is how we’ve been building them. For far too long, they’ve been oversized, bureaucratic and slow. That’s what needs to change.

Agile has always advocated smaller teams and elimination of wasteful tasks. We have specific meetings that are designed to deal with these very issues, called the retrospective. So now we have a way of implementing lean thinking and eliminating team waste.

AI can amplify the principles in the Agile manifesto rather than replace them by allowing teams to experiment, learn, and iterate at an unprecedented pace taking us into a new era of hyper agility.

In practice, this means Agile teams will become smaller, sharper, and more specialized. Instead of large cross-functional squads, we’ll see AI-augmented micro-teams, where a handful of people collaborate with AI copilots. A single Product Owner might work with an AI-driven backlog manager to help prioritize and keep the backlog healthy. A UX designer might iterate with a customer simulation AI agent that is able to mimic personas and customer behavior, catching usability issues early. The team remains; but it looks very different.

Could we move toward a world where one person, armed with AI agents, is able to work efficiently alone? Possibly. But the question isn’t just whether it’s possible, it’s whether it’s better and fulfilling.

Teams exist not just to get work done, but to challenge each other’s ideas, hold each other accountable, and push innovation forward. Teams will survive AI, but those that embrace AI as a collaborator, not a replacement, will be the ones that shape the future of work.

Find out how to get your teams up to speed with the latest Agile and AI training solutions.