Published: WMCH roundtable report on “Coll ective Intelligence vs Artificial Intelligence” – why we did it and what’s next

Dear all,

I’d like to share that Wikimedia CH and Open Future have published the 
report from our Lausanne roundtable “Collective Intelligence vs 
Artificial Intelligence” (4 Nov 2025), co-organised with Open Future and 
IMD Business School.

I also want to add a bit of context on /why/ we did this, because it 
comes out of a very concrete moment for many of us: over the past year, 
it has sometimes felt like “Wikimedia + AI” discussions were happening 
everywhere, yet at the same time the community’s shared space for 
sensemaking was getting thinner — less clarity, less common ground, and 
more uncertainty about where to focus limited time and resources. 
Internally at WMCH, we reached a point where we needed to step back and 
ask: *what should we do in this period, and where should we allocate our 
resources in a way that is useful for the Movement?*

 From that starting point, we began working with external partners to 
improve our understanding of what is changing in the information 
ecosystem. Among others, we exchanged with the AI research ecosystem in 
Switzerland (e.g., IDSIA - Dalle Molle Institute for Artificial 
Intelligence of Lugano) and, together with input from Open Future, we 
converged on a simple idea: instead of debating AI only through the lens 
of “editing with AI tools”, we should widen the circle and ask experts 
from outside the Movement what they see coming — and what it could mean 
for a knowledge commons like Wikimedia.

That led to the Lausanne roundtable: a deliberately “cross-ecosystem” 
discussion with people working in areas such as AI development and data 
science, journalism and media, research, and public-interest policy, 
alongside people from the Wikimedia ecosystem. The goal was not to 
produce a single answer, but to map the key tensions and update our 
collective assumptions about the near future — and to do it in a way 
that is transparent and shareable with the wider community.

The report summarises those insights and frames the emergence of a “new 
knowledge loop”, in which AI services increasingly become interfaces to 
knowledge. One concern raised is disintermediation: Wikimedia content 
can be heavily used by machines while fewer humans visit Wikimedia 
projects directly — with potential consequences for participation, 
feedback loops, and long-term sustainability. At the same time, the 
report argues there is no “going back”: AI influence is not avoidable, 
and the open question is how collective intelligence and AI can be 
combined while keeping Wikimedia human-centred.

A quick note on intent and framing: this is a /civic sensemaking/ 
effort. It is not a proposal to restrict open access for people, and it 
is not an attempt to tell the Movement what to do. The underlying 
question is: *how do we keep the commons open and accessible to 
everyone, while making its large-scale reuse sustainable and 
accountable?* In other words, how do we avoid a future where knowledge 
remains technically “open” but becomes practically concentrated behind a 
few AI interfaces?

This report is also meant as a first step. There is the plan to develop 
a draft white paper in 2026, building on these insights, to outline 
possible strategic directions and reactions — not only for WMCH, but as 
a contribution that other communities, affiliates, and interested 
Wikimedians can critique, improve, and build on.

Links:

  * Meta page (context + deliverables):
    https://meta.wikimedia.org/wiki/Wikimedia_CH/Innovation/CI_vs_AI#Deliverables
  * PDF on Commons:
    https://commons.wikimedia.org/wiki/File:Wikimedia_CH_REPORT_AI_v4_web.pdf

Feedback is very welcome.

-- 
Ilario Valdelli
Innovation programme lead
Wikimedia CH
Verein zur Förderung Freien Wissens
Association pour l’avancement des connaissances libre
Associazione per il sostegno alla conoscenza libera
http://www.wikimedia.ch

_______________________________________________
Wikimedia-l mailing list -- [email protected], guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
Public archives at https://lists.wikimedia.org/hyperkitty/list/[email protected]/message/GROBMQOEE3XXHW54B5KSPHKXD2W5F7VI/
To unsubscribe send an email to wikimedia-l-leave-RusutVdil2icGmH+5r0DM0B+6BGkLq7r@public.gmane.org