Is AI hurting team dynamics?

Our conversation revealed patterns that many of us are seeing across different organisations – patterns that suggest we need to be much more intentional about how we integrate these powerful new tools into our existing workflows.

The designer dilemma

One of the most immediate impacts we discussed was how AI prototyping tools are affecting already stretched design teams. Many organisations operate with designers serving multiple teams – a setup that was already problematic before AI entered the picture. Now, with product owners able to create functional prototypes using tools like Replit and Vercel, we’re seeing design expertise get bypassed entirely.

The pattern is becoming familiar: a product owner creates what looks like a sophisticated prototype, it gets prioritised into the next sprint and suddenly the designer is expected to provide detailed specifications for something they had no hand in creating. We’ve all seen the frustration this creates – designers feeling short-circuited out of the process while being held accountable for the quality of the final product.

When ‘done’ isn’t really done

We also explored a psychological challenge that many teams aren’t prepared for: AI-generated prototypes look finished. Unlike the hand-drawn wireframes or deliberately rough sketches that clearly signal work-in-progress, these tools produce polished interfaces that appear production-ready.

We drew parallels to the tailoring industry, where leaving loose threads on a suit during fittings psychologically signals to customers that changes are still welcome. AI prototypes don’t have those loose threads. They look complete, which makes stakeholders reluctant to suggest changes and teams less likely to question fundamental assumptions.

This polished appearance short-circuits the messy, divergent thinking that’s essential for breakthrough products. Instead of exploring problems, teams find themselves validating solutions that may have locked in assumptions too early in the process.

The expertise question

Our conversation touched on a fundamental tension: AI tools democratise capabilities that were previously restricted to specialists, but democratisation isn’t always beneficial. When anyone can create a functional prototype, the boundaries between product management, design and engineering blur – sometimes productively, but often in ways that undervalue specialised knowledge.

We discussed how product owners, empowered by these accessible tools, might take on design decisions they’re not equipped to make. Meanwhile, designers worry about their relevance as their expertise gets overlooked in favour of speed. The subtle knowledge that comes from years of experience in user research, information architecture and interaction design risks being lost when solutions can be generated quickly.

The accountability problem

One of the more philosophical questions we grappled with was whether AI tools should be treated as team members or remain firmly positioned as tools. Our consensus was clear: regardless of how sophisticated these systems become, they cannot be held accountable for outcomes. This creates a crucial gap where human judgment must bridge the difference between what AI can produce and what users actually need.

We’ve all seen teams struggle with this dynamic – the temptation to treat AI as a collaborator can lead to diffused responsibility and unclear decision-making processes. The most successful teams maintain clear human ownership while leveraging AI as an accelerator of human capability.

Practical patterns for better integration

Through our collective experiences, several patterns emerged for successfully integrating AI tools while maintaining healthy team dynamics:

Be explicit about purpose: Teams need clarity about whether they’re using AI for opportunity validation, solution exploration or detailed specification. Each purpose requires different approaches and levels of investment.

Honour your constraints: Rather than using AI to bypass team limitations, we should acknowledge our bottlenecks and optimise around them. If design capacity is your constraint, that should guide your pace of work, rather than circumventing it with AI shortcuts.

Maintain true collaboration: The most effective teams preserve cross-functional integration rather than allowing AI tools to fragment their collaborative processes. This means keeping designers, product managers and engineers working together from problem definition through solution delivery.

Plan for disposability: Teams that explicitly intend to throw away AI-generated prototypes after they serve their purpose avoid premature commitment to sub-optimal solutions. This requires discipline and clear communication with stakeholders.

The cultural dimension

Our discussion revealed that the impact of AI on team dynamics often reflects deeper organisational values. Companies that successfully navigate this transition tend to value process as much as output, expertise as much as speed and collaboration as much as individual productivity.

We observed that organisations where team members genuinely care about each other as individuals tend to handle these transitions better. They respect each other’s expertise, give space for different types of contributions and credit each other’s work appropriately. Where that care is absent – often due to environmental pressures rather than individual shortcomings – AI tools can exacerbate existing dysfunction.

The nuance factor

Throughout our conversation, we kept returning to the theme of nuance. The right approach depends on where you are in the product lifecycle, your team’s skill composition, the types of products you’re building and countless other contextual factors. There’s no one-size-fits-all solution, which is both the challenge and the opportunity.

This is where experienced practitioners add tremendous value – not in knowing the ‘right’ answer, but in helping teams navigate the specific complexities of their situation. The organisations that thrive will be those that invest in developing this contextual judgment rather than looking for universal solutions.

Looking ahead

We wrapped up by acknowledging that we’re still in the early stages of understanding how AI will reshape product development. The patterns we’re seeing today will likely evolve as tools become more sophisticated and teams develop better practices.

What seems clear is that the future belongs to teams that can harness AI’s power while preserving the essentially human elements of great product development: empathy, creativity and the ability to synthesise diverse perspectives into solutions that truly serve user needs.

The speed that AI provides is valuable, but only when applied with the wisdom to know what problems are worth solving and the patience to solve them well. As we continue to navigate this transition, the conversations become just as important as the tools themselves.