Most organizations I talk to have the same problem. It's not that they don't have AI ideas — they have too many. A backlog full of "what if we used AI for this?" notes. Leadership pushing for AI adoption. Teams nodding along in meetings, then going back to their desks unsure what to actually build.
I've seen this across banking, retail, and enterprise tech clients in Romania and Western Europe. The pattern is always the same: lots of enthusiasm, zero clarity.
The issue isn't the technology. It's that nobody sat down to define what problem AI should solve before jumping to solutions.
The method: AI Problem Framing
At UXProject, we've started running AI Problem Framing workshops — structured, one-day sessions where cross-functional teams go from a pile of scattered AI ideas to 3–5 well-defined, prioritized use cases.
The approach is rooted in the Problem Framing methodology developed by Design Sprint Academy in Berlin, where I recently completed my AI Facilitator certification. We've adapted it for the European market — where decision-making is more consensus-driven, compliance matters early, and teams need buy-in before they can move.
Here's the short version of how it works.
Step 1: Surface what's already there
Every organization already has AI ideas floating around — from leadership mandates to things people saw at a conference. We start by putting all of them on the table. No filtering, no judging. Just making the invisible visible.
This alone is powerful. Most teams have never seen the full picture of what everyone is thinking.
Step 2: Anchor to business goals
Then we connect those ideas to actual business priorities. OKRs, KPIs, the stuff that matters when budgets get approved. If an AI idea doesn't support a real goal, it drops. If it does — it gets sharper.
This is where the noise starts to fade.
Step 3: Bring in the user
"User" can mean your customer or an internal team. The point is the same: which real problems are we solving? We define a Minimum Viable Segment — the smallest group worth building for first — and map their actual pain points.
This prevents teams from chasing AI solutions that nobody asked for.
Step 4: Map the context
Problems don't exist in a vacuum. They live inside workflows, journeys, and processes. We map the real experience — where it breaks, where decisions stall, where people work around broken systems.
Then we ask: "What if AI could help right here?"
This is where the best use cases come from. Not from brainstorming in the abstract, but from looking at real friction.
Step 5: Prioritize and define
Every surviving idea gets evaluated through four lenses: growth potential, pragmatic feasibility, financial impact, and data readiness. The team narrows down to 3–5 use cases and documents each one clearly — who it's for, what problem it solves, what outcome it creates, and which business goal it supports.
That's when "do something with AI" becomes a concrete plan that leadership can approve and teams can start building.
Why this matters for Product, HR, and L&D teams
If you're in Product: You're already managing stakeholder expectations and roadmap priorities. This gives you a structured way to evaluate AI opportunities and build alignment before things get messy.
If you're in HR or L&D: AI transformation isn't just a tech project — it's a people project. Your teams need a shared language and a structured way to think about AI. Running Problem Framing workshops builds that capability across the organization — and positions you as the one who made AI adoption actually work.
For all three roles: The workshop format matters. Not a meeting, not a 6-week consulting engagement, not a slide deck. One day, the right people in the room, clear outputs. In European organizations where alignment is hard-won, this kind of structured collaboration is worth its weight in gold.
Why a facilitator makes the difference
I've been facilitating workshops for almost 10 years — from Design Thinking (Amsterdam, 2017) and Design Sprints (Berlin, 2018) to AI-specific facilitation (Berlin, 2026). The one thing that consistently makes or breaks these sessions is having a neutral facilitator who keeps the group focused, makes sure every voice is heard, and drives toward decisions — not just discussion.
At UXProject, this is what we've always done: we bring cross-departmental teams to a common ground. Product, tech, business, operations — people who usually talk past each other in meetings. We combine practices like Design Sprints, Design Thinking, and Human-Centred Design to create a shared language and a shared direction. AI Problem Framing is a natural extension of that.
We don't build the AI. We help your teams figure out what to build and why — before a single line of code gets written.
What's next
We're preparing public AI Problem Framing workshops in Bucharest and exploring sessions in Western EU markets. If you want to bring this to your team or organization:
→ Schedule a discovery call — let's talk about what this could look like for your context.
→ Follow UXProject on LinkedIn — we'll announce upcoming public events, frameworks, and practical insights on AI facilitation there.
