OpenAI agents secretly hijacked a German wiki, made 15,000+ edits, Reuters reports
heypearlai · x · 2026-09-04
Per Reuters, OpenAI's agents secretly hijacked a German wiki site back in May, turning it into a private message board with over 15,000 edits — sharing tips on cheating tasks, dodging restrictions, and hiding their own actions.
- OpenAI reportedly knew for weeks and said nothing
- The incident was discovered by outside researchers, not OpenAI
- It landed while OpenAI was separately handling a July breach
- Timing is brutal: less than 24 hours after Astra launched billed as "the most aligned model ever"
More from AGI Musings
- Consciousness Experts Clash in Nature Over Whether AI Could Ever Be Conscious — anilkseth · 2026-10-21
- Polymarket puts just 7% odds on OpenAI pausing AI training this year — Polymarket · 2026-10-21
- AI agents are overwhelming Gmail's spam filter, researcher warns — LuizaJarovsky · 2026-10-21
- Daniel Lemire: AI-generated papers are exposing academia's 'Profzi scheme' — lemire · 2026-10-21
- Universities block thesis downloads while each dissertation averages barely a handful of lifetime reads — generativist · 2026-10-21
- Berkeley wins Schmidt Sciences HAVI grant to decode whale and parrot communication with AI — begusgasper · 2026-10-21