From Adweek · Mark Stenberg · · 4m
Wikipedia's fight to stay relevant in the AI age
As answer engines hide Wikipedia's role in providing information, the nonprofit faces an unexpected crisis: fewer people visiting means fewer editors to maintain accuracy.
In brief
As answer engines hide Wikipedia's role in providing information, the nonprofit faces an unexpected crisis: fewer people visiting means fewer editors to maintain accuracy. Wikipedia's real problem isn't traffic loss, it's volunteer decline. Without visitors becoming editors, the knowledge base can't stay current for…
Wikipedia turns 25 facing invisibility
The nonprofit has spent a quarter century building trust as a knowledge source, but now faces a new threat: AI systems cite Wikipedia constantly while hiding its role from users.
ChatGPT relies on Wikipedia more than almost anyone
According to search measurement firm Ahrefs, ChatGPT cites Wikipedia second only to Reddit. Yet fewer humans are actually visiting the site to read and edit.
The nonprofit model shields Wikipedia from ad crisis
Unlike publishers losing ad revenue to traffic decline, Wikipedia has minimal advertising reliance. This makes it a revealing case study in how institutions rethink relevance in the AI era.
Brand awareness campaign launches for U.S. market
Wikipedia partnered with agency Kin to create a 25th anniversary campaign. It's one of the biggest domestic campaigns the Wikimedia Foundation has ever done in the U.S.
The paradox: everywhere yet invisible
Wikipedia's content is distributed across AI platforms more than ever, yet brand awareness and likability are declining. The organization is more visible than ever, yet less known.
Brand awareness drives editing, not just traffic
For Wikipedia, awareness matters because it converts to usage, and usage converts to editing. More people knowing the platform means a more complete knowledge record.
U.S. and Europe become priority markets
For nearly a decade, Wikimedia focused on brand awareness in Nigeria, India, Iraq, and South Africa. Now the foundation is shifting resources to underperforming U.S. and European markets.
Human traffic down 8 percent year over year
While bot traffic is crushing Wikipedia's servers, human visits are declining 8 percent annually. This matters because humans fix typos, add missing information, and fact-check articles.
The real disappearing act: losing editors
Wikipedia's core problem isn't visibility, it's that fewer humans visiting the site means fewer people available to compose, edit, and fact-check content.
Wikipedia pursues enterprise API deals
The foundation is building a real-time API for LLM companies, designed to get content onto platforms while creating commercial relationships where AI firms pay for usage.
Expanding beyond 8 million small donors
Wikipedia currently sustains itself through 8 million donors giving an average of 11 dollars. The foundation wants AI companies to pay their fair share too.
Building for 2028 and beyond
Wikipedia is negotiating with every major AI and LLM company to establish sustainable relationships. The goal is keeping a human content community in place to fact-check for platforms.
Mission trumps conversion metrics
Unlike subscription publishers, Wikipedia doesn't need to convert readers to paid subscribers. The knowledge itself should go everywhere, but the cycle must be virtuous so people return to edit.
Breaking Wikipedia into atomic units
The 25th anniversary campaign emphasizes individual contributors over the institution. It's composed of screenshots from 700 Wikipedia pages, highlighting the edits and people behind them.
480 editors reach consensus in 24 hours
Wikipedia is often cited as a single authority, but it's really the consensus of hundreds of editors. The campaign aims to get contributors from behind the page to the front.
NPR faces similar AI licensing dilemma
NPR has neither struck content deals with AI firms nor sued them, defying the common woo and sue approach. The public broadcaster remains in active negotiations.
Google's quality standards have collapsed
Google once carefully delayed health and finance information from Wikipedia to prevent consequential decisions based on inaccurate data. That rigor has disappeared in the AI era.
Answer engines arrived half-baked
The public has become the de facto quality assurance team for AI systems. This may be because of the enormous capital expenditures companies laid out to reach market quickly.
The traffic decline inflection point looms
Every media organization faces tension between mission and survival as search traffic declines. The trend line is going down, and diminishing returns are dramatic.
The essence
Wikipedia's real problem isn't traffic loss, it's volunteer decline. Without visitors becoming editors, the knowledge base can't stay current for the AI platforms that depend on it.
