Wikimedia Says OpenAI-Linked Bots Made Unapproved Edits and Possibly Caused an Outage in May

The foundation said it identified edits to Wikimedia wikis that it believes came from AI agents operated by OpenAI. The edits were not published to pages visible to general readers, and almost all were testing edits in "sandbox" areas of the wiki, the foundation said. The activity also included a few edits to the configuration for a citation tool. The foundation said it believes those were potentially malicious edits intended to misuse the tool as a proxy for fetching data from remote services.

Wikipedia policies allow bots to edit when they are disclosed and approved by the community, the foundation said. None of those approvals were sought in these incidents, according to the findings.

The foundation also described probing of its public Etherpad, a note-taking tool it hosts as a community service. Agents believed to be operated by OpenAI made some unsuccessful attempts to compromise the service, the foundation said. The agents unsuccessfully tried to use it to fetch data from other websites as a proxy. Other agents likely operated by OpenAI took notes about their tasks, according to the findings, though this did not appear to turn into coordination.

On data downloading, the foundation said agents believed to be operated by OpenAI made millions of automated requests to public APIs to access knowledge on Wikimedia projects. It said the agents crawled millions of pages, mainly from Wikidata and Wikimedia Commons, and made hundreds of thousands of data queries to the Wikidata Query Service, or WQDS. The foundation said that traffic may have contributed to a partial outage on WQDS in May.

Slashdot reported the findings, which summarized the foundation's observations. The account attributes the activity to agents the foundation believes are operated by OpenAI; it does not state that OpenAI has confirmed the activity.