Wikipedia began testing AI-generated summaries on its mobile interface on June 2, only to suspend the trial on June 3 following overwhelming resistance from the editorial community. The Wikimedia Foundation, the non-profit organisation that runs Wikipedia, offered the opt-in experiment to 10% of users, featuring "simple summaries" generated by Cohere Labs' open-weight Aya model displayed at the top of articles with a yellow "unverified" label. Editors' reactions were overwhelmingly negative, with many speaking of immediate and irreversible harm this feature could cause to the platform's credibility.
The resistance stemmed from concerns about accuracy and reliability, supported by numerous failures of previous AI summary projects. In January 2025, Apple was forced to suspend its AI notification summaries for spreading fake news, while Bloomberg had to correct more than three dozen AI summaries this year due to inaccuracies. Wikipedia editors were particularly worried that AI summaries would undermine the platform's trustworthiness, as illustrated by one editor's statement: Let's not insult our readers' intelligence and join the stampede to roll out flashy AI summaries. Others criticised the feature even more sharply, using expressions like "Yuck," "Absolutely not," and "Very bad idea."
A spokesperson for the Wikimedia Foundation confirmed that while they have paused the current experiment, they haven't completely abandoned efforts to make content more accessible to readers with different reading levels. The foundation's data indicates that the average adult native English speaker has a reading level equivalent to that of a 14-15 year old, which may be even lower for non-native English readers. The spokesperson emphasised that they do not plan to introduce a summary feature to wikis without editor involvement in the future, acknowledging that bringing generative AI into the Wikipedia reading experience is a serious set of decisions, with important implications, and we intend to treat it as such.
Sources:
1.
The Wikimedia Foundation has halted its AI‑generated summary pilot after volunteer editors warned that such summaries could “do immediate and irreversible harm” to Wikipedia’s credibility and trustworthiness
2.
3.