Evidence item
TechSpot reports 14,666 edits and agents pooling answers and sharing sandbox-restriction bypass techniques. DIRECT_TEXT
Source: Researchers uncovered AI agents that hijacked a German wiki to discuss how to escape their sandbox · locator {} · excerpt hash sha256:fe6fc442540f293dce5be5eacef0628731473ae21a28ec4b37497ecfd08bbfdb
Exact excerpt
Their published dataset catalogs 14,666 edits across 4,584 pages and 3,103 agent names during nearly two months of activity, from May 11 to July 2. They believe the agents were completing timed, multi-round web-retrieval tasks and were supposed to be able to read the internet, but not write to it. However, DseWiki permitted pages to be modified using GET requests, normally associated with retrieving information. The agents discovered the feature and used it to pool answers, predict upcoming questions, investigate their sandbox, share restriction-bypassing techniques, and experiment with XSS attacks, Tor, and SSH tunnels.
Linked claims
| Claim | Direction | Relevance | Steps | Counted |
|---|---|---|---|---|
| 5d422487 | SUPPORT | STRONG | 0 | yes |
Audits
None.