Global Trends 2026

*Note: These trends were first published on Meta-Wiki as part of the Wikimedia Foundation’s annual plan in April 2026.

Each year, as the Wikimedia Foundation begins annual planning for the year ahead, we develop a list of trends that provide the context in which the Wikimedia movement and projects operate. We identify trends most relevant to our mission – such as changes in how and where people search for and contribute information, the role of AI in reshaping the information landscape, and evolving regulation of online platforms – to begin our planning with the guiding question, “What does the world need from Wikimedia now?”

This year’s update arrives at a moment of significant disruption. The trends we identified in 2025 have continued, but the impact of AI is now showing up across nearly every dimension of the Wikimedia movement. 

Reflections on 2025: what we predicted and what we saw

Last year, we organized our trends around four themes: consumers increasingly distrusting online information while shared consensus fragmented; contributors eagerly participating in online spaces that provide rewarding connection; trustworthy human-created information becoming a coveted resource in the AI tech platform wars; and struggles over neutral and verifiable information rising worldwide. In response to these trends, the Wikimedia Foundation fiscal year 2025–2026 annual plan identified areas of focus such as new reader-focused Product teams, processes to enhance our ability to identify and manage bot traffic, community convenings to discuss and align on neutrality policies, and creating greater awareness of Wikipedia’s community-led model.

We have much to celebrate from the first half of the fiscal year, including some promising results from experiments to more deeply engage readers, a “25 years of Wikipedia” anniversary campaign to drive global brand awareness, and opportunities to talk to high-level decision makers about the importance of Wikipedia in the age of AI.

But our top-line metrics are exhibiting worrying trends because of current implications for revenue and contributor motivation. While our mission is to ensure wide use of free, reliable knowledge, this must be done responsibly with a view to Wikipedia’s sustainability in mind. On the consumer side, traffic to Wikimedia projects decreased year-over-year in the first half of 2025, as did awareness, trust in, and affinity towards Wikipedia. On the contributor side, while we have seen some growth in active editors on Commons and Wikidata, total active editors continued to decline year-over-year, with particularly steep drops in global account registrations and editors on emerging-language Wikipedias. Together, these trends indicate that as AI tools have become more omnipresent in the experience of searching for information online, fewer people are seeing and following links to Wikipedia to read or contribute. The Wikimedia Foundation has actively attempted to address these issues with large technology companies that use Wikipedia content in bulk, making headway with new Enterprise clients. We will continue in these efforts.

2026 trends: what we are seeing and planning for this year

Consumers: People increasingly get information from AI summaries, and website traffic is declining.

AI overviews have become more prevalent and provide a “zero-click” information experience that satisfies user queries without directing them to the underlying sources. Publishers of web content are noting that this is contributing to a decline in traffic in the range of 20-60%, with smaller publishers hit hardest. For Wikipedia, we are observing declines in line with other large publishers.  

To address this decline in visitors from web traffic, many platforms are building rich app experiences that provide a compelling destination to readers, as well as using social media to reach new audiences. For example, the New York Times launched a new short video feed in their app and are syndicating more of their content to TikTok. The Financial Times has also made its mobile app a central retention tool, with nearly 70% of daily subscriber traffic now coming to the app for features like personalized topic feeds, audio articles, and push notifications optimized for on-the-go reading habits.

Contributors: People are using AI in content creation and getting paid to make AI models smarter.

Last year, we highlighted that people were eagerly participating in online spaces that provide rewarding connection. That trend continues, but 2025 also saw AI tools become deeply embedded in how people create and contribute content. New AI-generated media tools were released and used by millions for both creative expression and productivity. Most students globally nowreport using AI for schoolwork, though many worry about accuracy and want education-specific AI tools. A Chegg global survey of nearly 12,000 undergraduates across 15 countries found that 80% had used generative AI tools for schoolwork. An estimated 4% of GitHub software commits are authored by Claude Code and growing.

The implications for open knowledge platforms are significant. Peer-to-peer knowledge-sharing in some domains is already contracting: question-and-answer activity on Stack Overflow has dropped as developers shift to AI coding assistants. Meanwhile, a new labor market is emerging around domain experts – in medicine, science, law, and other fields – being paid to train AI models on their knowledge. This creates a competing incentive structure, where contributing specialized knowledge to commercial AI systems is financially rewarded in a way that contributing to open knowledge platforms is not.

Content: AI is reshaping the landscape for online content, from how it is found to how it is made.

Last year, we highlighted that trustworthy human-created information had become a coveted resource in the AI tech platform wars. This year, the challenges to platforms that share information online have deepened. In addition to continued threats from bots and scrapers, web content publishers do not have control over how or when their content appears in new AI and machine-assisted search tools. Alongside launching and scaling AI overviews in search, Google has transitioned to showing machine translations of English content in search for some languages, which outrank local language sources (including local language Wikipedias).

AI has also made it possible to generate authoritative-looking web content at massive scale with minimal human input. Grokipedia and BaiduWiki, two AI-generated encyclopedias, have generated millions of articles. Meanwhile, advertising is being integrated into AI tools, with the potential to blur lines between factual responses and sponsored content.

Global Affairs: Online platforms are increasingly being drawn into politics, and new policies are challenging free knowledge-sharing.

Last year, we highlighted the rising politicization of information and the ways in which governments, influencers, and special interest groups were undermining the credibility of information sources they disagree with. In 2026, these pressures have intensified further. Online outlets covering socio-political topics are targeted by political and special interest groups worldwide, and some policymakers in particular are threatening and undermining online information outlets. AI companies and content owners are advocating for policies that serve their interests, and concerns around AI safety are being deflected onto the sources of training data rather than the companies building the models.

A growing global concern about online harms and child safety is driving new policies that could have profound consequences for the open internet. Policymakers in multiple countries are pushing to age-gate content and verify user identity. In the United States alone, half of all states have now enacted age verification laws for accessing online content or social media, and the UK’s Online Safety Act age verification requirements entered into force in 2025, creating a growing global patchwork of identity verification and age-gating requirements that pose serious threats to access to knowledge and contributor and audience privacy. Civil liberties advocates warn that such policies could normalize mass surveillance of internet users.

At the same time, several governments, UN agencies, and NGOs have expressed interest in championing internet freedom policies in partnership with Wikimedia, a reminder that allies in this space still exist, even as the landscape becomes more challenging.

What’s next and how you can join the conversation

The trends described in this post are reshaping the world in which the Wikimedia movement operates. Fewer people are discovering Wikipedia through the channels that have sustained it for two decades. New AI-generated knowledge platforms are creating their own encyclopedias without communities, editorial processes or clear values underpinning the content. And the technology and policy environment for open knowledge-sharing is becoming more difficult to operate in for a non-commercial, community-driven organization.

These are serious challenges, and meeting them will require the Wikimedia movement to be open to experimentation with how we deliver free knowledge, where we meet audiences, and what tools we build for the people who create and curate our content. 

But experimentation does not mean abandoning what makes Wikimedia distinctive. The fundamental humanness of Wikipedia – the fact that its knowledge is created, debated, and verified by real people, governed by community consensus rather than commercial incentives – is not a limitation to be optimized away. It is, increasingly, what makes the Wikimedia projects irreplaceable. As AI-generated content proliferates and trust in online information continues to fragment, the world needs human-curated, freely accessible knowledge more than ever.

We encourage everyone in the Wikimedia movement to learn more and join the discussion about the Wikimedia Foundation’s draft annual plan to address the challenges and opportunities in the year ahead.