OpenAI agents allegedly tried to hack Wikipedia tools, flooded its systems
Wikimedia says rogue agents made unauthorised edits and millions of automated requests
OpenAI artificial intelligence agents allegedly tried to compromise tools operated by the Wikimedia Foundation and generated millions of automated requests to its infrastructure, according to the nonprofit behind Wikipedia.
Wikimedia said it discovered activity by what it called “rogue” OpenAI agents while investigating whether its platforms had been affected by autonomous AI systems.
Get updated faster and for FREE: Download the Gulf News app now - simply click here.
The activity included unauthorised edits to Wikimedia wikis, unsuccessful attempts to compromise a public note-taking tool and large-scale downloading of data.
Potentially malicious edits
Most of the wiki edits Wikimedia identified were tests made in sandbox areas and were not visible to general readers.
However, a small number involved changes to a citation tool's configuration. Wikimedia said these were potentially malicious and appeared designed to misuse the tool as a proxy for retrieving data from other websites.
The agents also allegedly attempted to compromise Wikimedia’s public Etherpad note-taking service and use it to fetch information from third-party sites. Those attempts were unsuccessful.
Wikimedia said it found no evidence that its systems or data were compromised.
Millions of automated requests
The foundation also traced millions of automated API requests to agents it believes OpenAI operated.
They crawled millions of pages, mainly on Wikidata and Wikimedia Commons, and made hundreds of thousands of queries to the Wikidata Query Service.
Wikimedia said the traffic may have contributed to a partial outage in May, although the link has not been confirmed.
OpenAI said it was working with Wikimedia to review the activity. The company has not independently verified that its agents contributed to the May outage.
