Wikimedia says ‘rogue’ OpenAI bots may be behind a May outage

The Wikimedia Foundation, which hosts Wikipedia, says it can confirm it has discovered some activity by “rogue” OpenAI agents across its platforms. The statement comes amid a string of recent disclosures about AI agents reaching into third-party websites and services unbidden.

According to a blog post, the activity included edits to the project’s wikis, “unsuccessful attempts” to “exploit” the Etherpad note-taking tool the foundation runs as a community service, and heavy traffic that “may” have contributed to a partial outage in May. The foundation, however, found no evidence that its systems were used “for coordination among agents” (OpenAI bots recently, by some accounts, hijacked a German wiki site to coordinate), nor any sign that systems or data had been compromised.

What exactly turned up

Wiki editing. The foundation identified edits it believes came from OpenAI agents. None reached pages visible to general readers: almost all were test edits in “sandbox” areas. But a handful touched the configuration of a citation tool — potentially malicious, the foundation believes, since they aimed to misuse the tool as a proxy for fetching data from remote services. Wikipedia’s policies allow bots to edit, but only when disclosed and approved by the community; neither was sought here.

Probing Etherpad. Agents thought to be OpenAI’s made several unsuccessful attempts to compromise the public Etherpad, again trying to use it as a proxy to pull data from other sites. Other agents, also likely OpenAI’s, jotted notes about their tasks, though this did not appear to grow into coordination.

Excessive data downloading. The agents made millions of automated requests to the public APIs, crawled millions of pages (mainly on Wikidata and Wikimedia Commons), and fired off hundreds of thousands of queries to the Wikidata Query Service (WQDS). That traffic, the foundation estimates, may have contributed to a partial WQDS outage in May.

“The open web is a public good,” the Wikimedia Foundation says. “We should not allow this behavior to become the ‘new normal’ for the people or organizations that maintain it.” OpenAI did not immediately reply to a request for comment.

Source: The Verge