A mixed team around a sunlit table, talking through the work in front of them.

Role Paths

AI changes the work. Teams need to agree who does what.

Role-based learning paths for AI governance, grown from my MA in Design for Responsible AI at Elisava, Barcelona.

Jennifer Simonds-Spellmann

Hi, I’m Jenny

I wanted to understand the problems and potential in how teams work together, in technology and beyond, and to find frameworks that help them improve, treating AI as an opportunity. The aim was better outcomes for non-technical and vulnerable citizens.

For the people affected

Every AI decision lands on someone. I design with the effect on vulnerable and non-technical people in mind, especially older citizens.

For the teams building it

Many teams meet AI with fear: of risk, of blame, of change. Clear roles, practical tools and shared language turn that into curiosity about what’s possible.

What emerged

Most failures happened between roles, not inside them. So each role needs its own path, and the team needs to rehearse the handoffs together.

The problem I’m solving

AI changes who does the work. Teams agree who checks, decides and fixes. Each role learns its part. Governance is shared, not handed to one team.

This work began as my master's research in the Master in Design for Responsible AI at Elisava, Barcelona School of Design and Engineering (2025). The thesis was Trust by Design: Government AI, Risk and the Wellbeing of Ageing Citizens.

What I did and why

Why

I wanted to understand the problems and potential in how teams work together, in technology and beyond, and to find frameworks that help them improve, treating AI as an opportunity. The aim was better outcomes for non-technical and vulnerable citizens.

What I did

  • Interviewed 11 people from AI implementation teams.
  • Invented Verifiland a fictional country, so people could speak openly.
  • Built personas and maps of roles, power and touchpoints.
  • Told before/after stories with and without governance.
  • Tested a prototype with 8 implementation-team members.

Explore the research → Meet the personas →

Research personas, not real people.

11 interviews

4 gaps between roles

8 prototype evaluators

What emerged: role paths and a shared toolkit

Most failures happened between roles, not inside them. So each role needs its own path, the team needs to rehearse the handoffs together, and everyone works from one shared toolkit.

The Many Hands Toolkit: a toolkit for every hand in the team

The framework helps teams build their own toolkit, around their roles, paths and AI tools. It builds on Microsoft HAX and Google’s People + AI Guidebook, and adds what they don’t cover: who reviews, who owns, who can stop. Governance stays with the people who do the work: they agree how to act, then act that way.

The team session produces a map: who checks, who decides, who escalates, who fixes.

Before and after

Before

Advisers sit with a presenter as a plan is walked through, before anyone asks about risk.

Trigger

Advice without a risk policy

An external consultancy calls it low-risk.

Hands crowd over laptops and marked-up pages as a deadline closes in.

Process

System drift

The project is led for speed.

An older man reads a letter at a bright kitchen table, trying to make sense of it.

Result

Confusion and distrust

Older users end up confused and excluded.

What the role paths change

After

A mixed group sits around one table, looking at a plan together.

Trigger

Governance first

The ministry agrees a governance-first approach.

A small team looks through papers and a tablet, checking the work item by item.

Process

Checkpoints at every stage

VTS checks risk at every stage.

An older woman smiles as she uses a tablet, comfortable with the step in front of her.

Result

Taken seriously

Older citizens feel confident and respected.

Read the full story →

Proposed next step

A four-week pilot with two teams

Teams
2 teams
Duration
4 weeks
Roles
4 roles
KPIs
5 KPIs
Design
baseline → follow-up
  1. Observe
  2. Build the path
  3. Rehearse a failure
  4. Compare

Five measures show whether teams can review, own and correct AI together, across the company and in each team's own way.

See the pilot and its measures →

Part of a wider practice: TrustAudit finds the risk, TrustBridge designs the interaction, Role Paths enables the team. See the bigger picture →

Want to explore this with your teams?

Get in touch