Nupco’s Graduate Development Program — Training the next generation of Saudi AI innovators

The program

In partnership with Nupco — Saudi Arabia's National Unified Procurement Company, running a national initiative to build innovation capability in the Kingdom's next generation — Design Sprint Academy delivered AI Problem Framing and Design Sprint training across five universities:

King Saud University and Princess Nourah Bint Abdulrahman University in Riyadh, King Abdulaziz University in Jeddah, Imam Mohammad Ibn Saud Islamic University in Riyadh, and King Fahd University of Petroleum and Minerals in Dammam.

Over 150 students. Multiple cities. A single goal: give the next generation of Saudi innovators a structured method for thinking with AI — not just using it.

^{Nupco captured these moments in a powerful video — a glimpse of the energy, creativity, and ambition shaping the next generation of innovators in the Kingdom.}

What we expected vs. what we found

When you train corporate teams in AI Problem Framing, there's a predictable dynamic in the room. People arrive with existing mental models of what their organization does, what their role permits them to question, and what kinds of ideas are realistic versus naive. Those mental models are often correct and useful. They're also constraints.

With students, none of that applies.

No existing workflows to protect. No organizational politics shaping which ideas are safe to propose. No prior investment in a particular solution direction that makes questioning the premise feel threatening. Students encounter the AI Problem Framing process without the accumulated scar tissue of corporate experience — and what comes out of that encounter is different in character from what experienced teams typically produce.

This isn't a romantic claim about beginner's mind. It's a specific observation about what the absence of organizational constraint produces when a structured methodology is applied.

What AI Problem Framing develops — and why it matters for this audience

AI Problem Framing is a structured process for identifying AI use cases that are worth building — grounded in real human needs, viable within organizational constraints, and specific enough to be tested before significant development resources are committed.

The core discipline it develops is the same one that determines the quality of AI adoption in organizations: the capacity to define the right problem before proposing an AI solution. Most AI adoption failures aren't technology failures. They're problem definition failures — situations where a technically capable solution was built for a poorly understood need, or where the AI was fitted to the most visible symptom rather than the most important underlying challenge.

For students who will enter organizations already deploying AI at scale, this is the capability that will determine their usefulness. The organizations they join will have data scientists, engineers, and AI tooling. What they will often lack is people who know how to frame the right problems for those capabilities to be applied to. That's what this training developed.

What the students produced

In a few hours of structured work, the student teams produced AI use case concepts that reflected genuine problem definition — not technology-first thinking. Four in particular are worth naming because they illustrate what the methodology produces when constraint is removed.

AI-powered navigation assistance for visually impaired people. Starting from the user's actual experience — the specific conditions that make navigation difficult — rather than from a technology capability seeking a problem to solve.

An AI advisor for students navigating university life. Framed around the specific decision moments where students are underserved by existing resources: choosing modules, managing academic pressure, finding support. Not a generic chatbot, but a targeted response to a defined gap.

A facility management assistant for campus operations. Identifying the operational friction points that campus staff actually experience — the problems that consume time and attention without clear systems for resolution — and proposing an AI layer that addresses the specific bottlenecks.

A tool to unify loyalty points across airlines, hotels, and credit cards. Built from the consumer frustration of managing fragmented reward systems, and framed around the specific moments where that fragmentation creates the most friction.

Every one of these concepts began with a customer need, was bounded by feasibility considerations, and was specific enough to describe what a prototype would need to demonstrate. That's not what happens when people are told to "think of AI ideas." That's what happens when people are given a structured process for defining problems before proposing solutions.

The observation that matters for corporate AI adoption

Watching 150 students work through AI Problem Framing confirmed something that we've seen in corporate training contexts but that is harder to isolate when organizational factors are also present: the bottleneck in AI adoption is almost never technical capacity. It's problem definition capacity.

In corporate environments, this bottleneck is obscured by other factors — organizational politics, existing solution investments, risk aversion, stakeholder complexity. These are real and legitimate constraints. But they often prevent teams from seeing that the underlying issue isn't the AI tools available, or the engineers to build with them, or even the budget to invest. The issue is the absence of a structured method for defining what's actually worth building.

Students with no corporate experience and no AI background, given that structured method for a few hours, produced better-framed use case concepts than many experienced teams produce after months of AI strategy work. That's not a critique of those teams. It's an observation about what structure does when it's applied without the accumulated weight of organizational constraint.

The implication for organizations investing in AI capability is direct: the return on training in how to frame problems well is higher than the return on training in how to use AI tools better. Tools change constantly. The capacity to ask the right question about what a tool should be applied to is the durable capability.

What the Nupco program represents

The partnership with Nupco was not a standard corporate training engagement. It was a national capability-building initiative — an investment in a generation rather than a team.

That framing matters. Vision 2030 sets an ambitious agenda for Saudi Arabia's digital transformation. The ambition is real. What determines whether it compounds over time is the capability of the people who will execute it — not just to use AI tools, but to make the decisions about what those tools should be used for.

The Nupco program is an investment in exactly that capability. Training 150 students in AI Problem Framing and Design Sprint methodology across five universities in multiple cities is not a symbolic gesture. It's a practical investment in the problem definition skills that will determine the quality of AI adoption decisions across organizations that these students will join and eventually lead.

For Design Sprint Academy, being a training partner in that program is significant. The methodology we've applied in corporate settings with organizations like ELM Company, Autodesk, Blue Shield of California, and NHS England transfers directly to an educational context — because the underlying challenge is the same wherever it appears. Too many solutions looking for problems. Not enough structured thinking about which problems are actually worth solving.

A note on energy

One thing that doesn't transfer from a written account of this program is the quality of engagement in the room. 150 students, working late into the evening, on problems they'd chosen because they found them genuinely interesting. The energy didn't flag. The ideas kept developing. The collaborative instinct — something we've observed consistently in Saudi teams, and wrote about in our facilitation guide for Saudi Arabia — was present throughout.

That energy, combined with a structured method, is the combination that produces the output described above. Energy without method produces enthusiasm. Method without energy produces process. Both together produce something worth paying attention to.