The Real Value of Collaborative AI Design

A clear theme has emerged across conversations I've had with organisations of all sizes and sectors this year: teams are struggling to see the strategic benefit from AI because teams rarely get the chance to step back from 'Business as Usual' and think together about how AI could meaningfully support their work.

When those conditions are created - structured time, shared attention, and a practical problem space - teams can move past abstract conversations about “using AI more” and into concrete workflows, decisions, and prototypes that impact the way their team operates.

After running a high energy AI Hackathon this week, here are some observations.

1. The Importance of Protected Time

One of the most stubborn barriers to effective AI adoption is the absence of protected time to experiment properly.

When teams only interact with AI tools in the gaps between meetings, the learning curve gets broken. People try something quickly, get haphazard or shallow results, and naturally form assumptions about the technology’s limits. Over time, this creates a predictable pattern:

• inexperience leads to...

• inconsistent outcomes, which leads to...

• Frustration or distrust, which makes people...

• Less likely to experiment again.

Protected time interrupts that cycle.

When people finally get space to focus as a team, they build a more accurate mental model of what AI can and can’t do; and they can compare and refine their thinking with others instead of working in isolation.

That shift from hurried, individual trial-and-error to deliberate, collective exploration is often the point at which teams stop treating AI as a bolt-on (or anxiety) and start understanding how it might genuinely support their work.

2. Collaboration is where alignment and knowledge-sharing happen

In our sessions, we could accelerate whole sections.... by using AI. Tools can generate workflows, outline steps, or propose structures very quickly.

But the process of teams exploring those workflows together - debating the logic, understanding the dependencies, and challenging the assumptions - is where the shared mental model forms.

You can’t automate or augment teamwork without alignment on how that work should get done, and on where human value lies.

3. User testing in the session is hugely valuable (even though the team is testing on themselves!)

When teams swap early versions of their ideas, several productive things happen:

• Weaknesses surface early, and blind spots become visible.

• People contribute to and strengthen each other’s thinking.

• Ownership becomes distributed, not isolated to whoever “built” it.

It’s a simple practice, but it speeds up learning and helps ideas travel across the organisation. (It's also a lot of fun!)

4.  Not all perceived value shows up as “speed”

A common way people evaluate an AI solution is by asking, “Is this faster than doing it myself?”. Speed is a massive superpower of AI but it’s only one dimension of value.

✅ Externalising tacit knowledge

Turning expert reasoning into explicit steps makes it accessible to anyone involved in the work.

✅ Supporting shared understanding and alignment

When the underlying decision logic is encoded, everyone works from the same assumptions rather than individual interpretations.

✅ Reducing variability caused by human inconsistency

Not because the AI is inherently consistent - it obviously isn’t -

but because a structured workflow gives both humans and AI a clearer standard to follow.

✅ Providing a foundation for iterative improvement

Once the logic is visible and explicit, it becomes much easier to refine, extend, or automate over time.

✅ Reducing reliance on single individuals

Workflows become more resilient when the reasoning isn’t locked in one person’s head.

✅ Strengthening an organisation’s collective intelligence

AI-enabled workflows don’t just codify internal processes, they make it possible to build and maintain richer knowledge repositories: patterns in client behaviour, recurring market dynamics, decision rationales, contextual insights, and the accumulated experience that normally disperses when people move roles or leave the organisation.

These repositories become a strategic asset in their own right. They help teams make better decisions, identify opportunities earlier, and create more tailored solutions for clients or customers. And because they’re continually updated through use, they form the basis of systems that learn and adapt over time.

Closing thoughts

AI-build sessions are not simply technical exercises; they are exercises in alignment, articulation, and shared reasoning.

When teams are given structure, language, and time to think together, the AI systems they produce are more robust, more usable, and more likely to survive contact with real work.

If your organisation wants to explore AI in a way that supports real work, accelerates learning, and builds a foundation for adaptive systems, collaborative sessions like these are a practical way to get started.

Drop me a line if you’d like help designing a version that suits your culture and pace, I’m always happy to support.