Individual skills & posture
Developing the human capabilities that support responsible engagement with AI. This includes AI literacy and the 4D framework (Describe, Discern, Delegate, Diligence).
The collective
Varied expertise, one shared ambition
A collective of complementary expertise united around a shared ambition: supporting impact organisations towards AI governance with the ethical excellence they deserve.
Our mission
Responsible AI in service of impact organisations
Supporting organisations working for the common good to develop informed AI governance: rooted in their identity and values, grounded in a rigorous and precise understanding of what the technology can (and cannot) do, and supported by human skills, adapted technical infrastructure and a lasting collective dynamic.
The foundations of responsible adoption
Informed AI Governance
Our approach rests on the articulation of three dimensions that together constitute informed AI governance. Hover over each circle to explore its definition.
- Identity & Mission — Knowing your mandate, your values and the context in which you operate.
- AI Capabilities — Understanding what AI can do and how to make the most of it.
- Managing AI risks — Identifying limitations, biases and unintended effects.
- Informed Governance — The intersection of all three dimensions for sound, sustainable decisions.
Our manifesto
Promoting AI in service of the collective good
Human-Centered AI places the needs, values and well-being of people at the heart of AI development and deployment. It is not about the technology itself, but about what it does for people.
Any AI technology we promote, develop or integrate into our training programmes must be designed to strengthen human capabilities — not to replace or bypass them.
We also draw on Ivan Illich's criteria for evaluating whether a tool can be considered just and "convivial". These criteria apply fully to AI.
It does not undermine personal autonomy
The tool remains at the service of the person. It does not create dependency, capture attention, or replace judgement.
It creates neither slave nor master
No forced subordination, no exercised domination. Power relations must not be reproduced — neither between humans and machines, nor between organisations.
It expands the radius of personal action
It gives everyone more capacity, more freedom of action, more agency in the world. AI must be emancipatory.
Enabling conditions
Three conditions for lasting adoption
These are not skills to acquire once, but dimensions to cultivate over time — at the level of a team, an organisation, or the sector as a whole.
Environment & information systems
Understanding the technical context in which any AI project takes place: information systems, infrastructure, implementation conditions and technical requirements.
Collective dynamics & shared learning
Creating a favourable social environment: collaborations, joint projects, sharing governance frameworks, and peer-to-peer experience exchange.
WHAT SETS US APART
Our approach in four principles
Posture: A compass, not a manual
We don't offer ready-made recipes. We develop in each organisation the capacity to navigate on its own — with discernment, in line with its values and mandate.
Technical rigour: Mastery in service of meaning
We know the technology from the inside — its architectures, how it works, its blind spots. This engineering rigour allows us to accompany without overselling, and to identify what AI can genuinely do for your mission.
Ethics: For AI as a common good
Every use we promote is assessed against a simple principle: does it strengthen the autonomy of people and organisations, or does it create dependency? AI must emancipate, never subjugate.
Collective: Intelligence built together
Responsible AI adoption does not happen organisation by organisation. It is built through peer exchange, the sharing of common frameworks, and the gradual emergence of a shared sectoral culture.