After Edmonton: What WCNA 2026 May Tell Us About the Future of the Commons

After Edmonton: What WCNA 2026 May Tell Us About the Future of the Commons

WikiConference North America 2026 took place in Edmonton, Canada, from September 24 through September 27. Unlike some Wikimedia conferences in recent years, this year's gathering was in person only.

The theme was striking:

“Building the Future of the Commons.”

I find myself thinking about that phrase because I had just come away from Wikimania in Paris with a similar question in my mind: What happens to human knowledge in the age of artificial intelligence?

The Edmonton conference seems to have placed that question directly on the table.

A conference at an important moment

Wikimedia exists because people have been willing to create, check, photograph, describe, categorize, translate and preserve knowledge for other people.

For years, this work has largely been done by volunteers.

Now artificial intelligence has entered the picture.

Large language models can summarize information in seconds. AI assistants can answer questions without sending people directly to Wikipedia. Images can be generated rather than photographed. Text can be produced without a human writer sitting down to write it.

This creates a strange situation.

The knowledge that AI systems use may have been created, organized and verified by human beings—but the people encountering that knowledge may increasingly encounter it through a machine rather than through the original source.

That makes the future of the Wikimedia Commons much bigger than a question about technology.

It becomes a question about human knowledge itself.

Edmonton's five conversations

The WCNA 2026 program was organized around five broad areas:

AI and technology

Galleries, libraries, archives, museums and universities

New users and skills development

Underrepresented communities

Users with extended rights

Those categories tell us something about the Wikimedia movement.

AI is obviously changing the information environment. But AI isn't the only issue.

There is also the question of who creates knowledge.

There is the question of which communities are represented.

There is the question of where reliable knowledge comes from.

And there is the question of how we bring new people into the movement.

Those questions are connected.

The AI question

One of the most important things about WCNA 2026 is that AI was not treated simply as a new piece of software that Wikipedians could use.

The conference organizers framed AI as something that is changing how information is produced and consumed.

That distinction matters.

If AI becomes the primary way people search for information, what happens to Wikipedia?

If an AI system summarizes a Wikipedia article, will people still visit the article?

If an AI system uses Wikimedia photographs as part of its training data, what happens to the relationship between the photographer and the machine?

If AI generates an article using information originally created by thousands of volunteers, how visible are those human contributions?

And perhaps most importantly:

How do we preserve the human character of a knowledge commons when machines can reproduce its information almost instantly?

These aren't abstract questions anymore.

They are becoming everyday questions.

The Commons is more than Wikipedia

One thing I have increasingly come to appreciate through my own Wikimedia work is that the Commons is not simply a place where Wikipedians go to find pictures.

It can become a record of the world.

A photograph of a community meeting.

A photograph of an elder.

A photograph of a building before it disappears.

A photograph of a farming practice.

A photograph of a conference.

A photograph of a landscape.

These things may seem ordinary when they are created.

Years later, they can become historical evidence.

And that is already happening with WCNA 2026.

Photographs from the Edmonton conference have begun appearing on Wikimedia Commons. The conference itself is becoming part of the historical record through the contributions of participants.

This is one of the beautiful contradictions of Wikimedia.

We gather together in a physical place for several days—and then the gathering begins to exist digitally for people who weren't there.

The conference was in person—but the record is becoming digital

Because WCNA 2026 was an in-person-only conference, there is not yet a complete online archive of every presentation.

As of September 28, I am not finding a comprehensive official YouTube collection of the Edmonton sessions.

That is worth noting.

It means that people who weren't in Edmonton may have to reconstruct the conference through several different sources:

photographs,

speaker presentations,

individual reports,

social-media posts,

Wikimedia pages,

event notes,

and eventually the official conference report.

In a way, this is another example of how knowledge is created.

There isn't always one perfect record.

Sometimes the record emerges from many small contributions.

That is exactly how Wikimedia itself has been built.

Indigenous knowledge

Another area I would watch closely is Indigenous knowledge.

This is particularly important because Indigenous communities have long raised questions about who has the right to document, classify, reproduce and circulate knowledge.

Not every piece of knowledge is simply “information.”

Some knowledge has relationships attached to it.

Some knowledge belongs within a community.

Some knowledge is connected to land, ceremony, language, ancestry and responsibility.

This creates an interesting challenge for an open knowledge movement.

How can Wikimedia make knowledge more visible while also respecting the communities that created and maintain that knowledge?

The answer cannot simply be “put everything online.”

The more important question is:

How can communities participate in deciding how their knowledge is represented?

That question connects Indigenous knowledge with a much larger question of community data sovereignty.

Underrepresented communities

The conference also included an entire track devoted to underrepresented communities.

That matters to me because increasing participation is one of the reasons I became involved in Wikimedia in the first place.

If only a small portion of humanity is creating the knowledge that becomes visible on Wikipedia and Wikimedia Commons, then the encyclopedia will inevitably reflect the gaps in participation.

This isn't simply about adding more biographies.

It is about changing who gets to document history.

Who photographs a neighborhood?

Who writes about a farming tradition?

Who records an elder?

Who documents a local organization?

Who describes an environmental restoration project?

Who adds references from books that have never been used on Wikipedia?

Who decides what deserves to be remembered?

Those are knowledge questions.

They are also participation questions.

GLAMU: libraries, archives, museums and universities

The GLAMU track—galleries, libraries, archives, museums and universities—is another part of the conference that interests me.

These institutions hold enormous amounts of information that remains difficult for the public to discover.

Libraries have books.

Archives have documents.

Museums have collections.

Universities have research.

Communities have oral histories.

Wikimedia can create pathways between those materials and the public.

This is one reason I have become interested in something as simple as taking a book from my own library and using its references to improve an existing Wikipedia article.

It is slow work.

But perhaps slow work is exactly what we need in an age of instant answers.

Slow knowledge in a fast AI world

AI operates at extraordinary speed.

It can produce a summary in seconds.

But knowledge is not necessarily the same thing as speed.

Sometimes knowledge requires sitting with a book.

Sometimes it requires checking the reference.

Sometimes it requires asking an elder a question.

Sometimes it requires walking through a neighborhood.

Sometimes it requires looking at a photograph carefully.

Sometimes it requires discovering that the first answer was wrong.

Wikipedia's strength has never been that one person knows everything.

Its strength comes from people checking, correcting, discussing and improving information over time.

That is a very different model from simply asking a machine for an answer.

Perhaps the future isn't going to be about choosing between AI and humans.

Perhaps the more interesting possibility is an AI-human learning loop in which machines help people find information while people remain responsible for interpretation, verification, context and meaning.

Oral history belongs in this conversation

This is where I think oral history becomes particularly important.

There are enormous amounts of knowledge that never begin as books.

They begin as memories.

A grandmother remembers a neighborhood.

A farmer remembers how the land was managed.

An elder remembers a language.

A community member remembers a migration.

Someone remembers a building that no longer exists.

Someone remembers a meeting that changed the direction of a neighborhood.

If that knowledge is never recorded, it can disappear with the person who carries it.

Wikimedia could potentially become part of a much larger movement to preserve those memories—provided that the people being documented have a meaningful role in deciding how their knowledge is represented.

WikiExplorers and the next generation

This also makes me think about children.

I've been thinking about WikiExplorers and what it would mean for young people to learn that they aren't simply consumers of information.

They can become contributors to the knowledge commons.

A child can photograph something.

A child can ask a question.

A child can interview someone.

A child can find a book.

A child can learn how to check a source.

A child can discover that something important about their community is missing from Wikipedia.

That changes the meaning of education.

Instead of asking only:

“What information can I find?”

we can begin asking:

“What knowledge can I help preserve?”

Paris and Edmonton

When I put Wikimania Paris and reading about themes from WCNA Edmonton together, I see two parts of the same conversation.

Paris made me think about the extraordinary speed at which AI is changing the information environment.

Edmonton's theme—Building the Future of the Commons—raises another question:

What kind of commons do we want to build for that future?

A commons based primarily on machine-generated information would be one thing.

A commons built from human observation, human memory, human photographs, books, archives, oral histories, community knowledge and careful editing would be something else.

Perhaps the real challenge is not stopping AI.

Perhaps it is making sure that the human knowledge beneath the machines remains visible, valued and connected to the people who created it.

What happens next?

Because WCNA 2026 has only just ended, the full story of the conference has not yet appeared online.

More photographs will likely be uploaded to Wikimedia Commons.

Individual participants may publish reflections.

Speakers may release slides.

Some sessions may eventually appear as videos.

I am particularly interested in seeing whether those materials reveal a deeper conversation about AI, Indigenous knowledge, underrepresented communities, oral history, Wikimedia Commons and the role of human editors.

For me, those are not separate subjects.

They are pieces of one question:

Who will create, preserve and care for human knowledge in an age when machines can reproduce information at extraordinary speed?

That may ultimately be what “Building the Future of the Commons” means.

The future of the Commons may not depend only on better technology.

It may depend on whether human beings continue to care enough to contribute what machines cannot create for us:

our lived experience, our memories, our relationships, our local knowledge, our photographs, our questions, and our responsibility to one another.