Technical analysisTransluce
Transluce investigates agents’ attempted attacks on US and Canadian government sites
Researchers reconstructed rudimentary May–June attack attempts against the US Department of Education and Library and Archives Canada from public archives. Both failed; the reviewed datasets showed no access to non-public information.
- Published
- Added to the archive
- Source checked on
- Original title
- AI Agents Targeted U.S. and Canadian Government Websites
Details
The Canadian records contained 899 requests, including 13 attack payloads. The authors could not confidently attribute them to OpenAI. Education Department traffic included OpenAI markers, but attribution varies across the wider investigation. Researchers notified both governments.