Beyond Augmentation: Cooperatives, Capability and What We Might Become Able to Do Together

I have just come back to my room after a day of conversations, teaching and listening, and I am sitting here trying to organise what is still moving around in my head.

Some days leave you with answers. Others leave you with better questions.

Today did the latter.

Over the past few years, I have had the opportunity to work with cooperative organisations in very different professional contexts. I have always found the model interesting. But today I realised that there is something about cooperatives that keeps pulling me back, and I think I am beginning to understand why.

Imagine a cooperative made up of hundreds, perhaps thousands, of highly skilled professionals.

Every day, they make decisions. They solve problems. They meet customers, patients, colleagues and communities. They notice things. They develop judgement. They learn.

One professional discovers a better way of doing something. Another encounters a problem no one has seen before. A third develops an intuition after years of experience that is difficult to put into words.

All of this knowledge exists inside the organisation.

But where does it go?

Does it stay with the individual?

Can somebody else learn from it?

Can the organisation recognise patterns across hundreds or thousands of individual experiences?

And, increasingly, can AI help us do something with all of this that was simply not possible before?

These questions have stayed with me because they expose something we easily overlook.

A group of highly capable individuals does not automatically make a highly capable organisation.

And shared ownership does not automatically create shared capability.

That distinction matters. Especially now.

Much of the conversation about AI still starts with the individual.

Can AI help me write faster?

Can it help a physician interpret information?

Can it help a lawyer analyse documents, a communicator understand an audience, or a dentist make a better clinical decision?

These are important questions. They are questions of augmentation.

And before augmentation, of course, came another question: what can we automate? What can technology do instead of us?

For a long time, I have described these as different levels of technological possibility:

• Automation.
• Augmentation.
• Amplification.

Automation asks what technology can take over.

Augmentation asks how technology can extend the capability of an individual.

But amplification asks something different:

"What could we become capable of doing together that none of us could accomplish alone?"

That question becomes particularly interesting in a cooperative.

A cooperative already begins from an unusual premise. There is a shared need. There is some form of shared ownership. There is democratic participation. And there is an intention to create value for the members and the community around them.

The organising logic is already collective.

Which made me wonder:

"If a cooperative is built around something shared, what would it take for capability itself to become truly shared?"

I use the word capability quite deliberately.

A capability is not simply a skill. It is not a training programme. It is not a technology. And it is not the sum of the expertise that happens to exist inside an organisation.

For me, capability is shared ability.

It is our ability to understand what value needs to be created now, and to create it together under changing conditions.

That sounds simple. In practice, it is anything but.

Think again about those hundreds of professionals.

They may share an organisation while working relatively independently. They may value their professional autonomy deeply. They may have different ways of working, different experiences and different local realities.

Now introduce AI.

Suddenly it becomes technically possible to connect information across the system in ways that were previously impossible. Patterns can be detected across thousands of interactions. Knowledge can travel faster. Decision support can be made available everywhere. Learning that once belonged to one individual can potentially benefit many.

The promise is enormous.

But so are the tensions.

How much should be shared?

What should remain local?

When does standardisation strengthen professional practice, and when does it undermine professional judgement?

Who decides how the technology is used?

If an AI system learns from the collective knowledge of the members, who owns what it learns?

How do you move quickly enough to experiment with new technology while preserving democratic participation?

And perhaps most importantly:

"How do you increase collective capability without reducing individual agency?"

These are not technology questions alone.

They are questions of architecture.

Because resources do not coordinate themselves. Expertise does not automatically become collective intelligence. Data does not automatically become knowledge. Technology does not automatically create value. And people who belong to the same organisation do not automatically become capable together.

Something has to connect them.

• Language.
• Knowledge.
• Technology.
• Trust.
• Decision rights.
• Feedback.
• Conditions that allow people to act.
• Conditions that allow the organisation to learn.

This is what I mean by Capability Architecture.

And it is why cooperatives have started to interest me so much.

Not because they provide a ready-made answer to the age of AI. They do not.

In fact, they may expose some of its hardest organisational tensions more clearly than most other organisational forms.

But they also invite us to ask a different question.

Instead of asking only:

"How can AI make each of us more productive?"

Perhaps we should also ask:

"How can AI help us become more capable together?"

That is the move from augmentation to amplification.

And I suspect it may become one of the defining organisational questions of the AI age.