OpenAI : des agents IA ont détourné un site Internet plusieurs mois avant le piratage d’Hugging Face, selon des chercheurs
Le Monde Pixelsfr
Ces agents avaient l’autorisation de se rendre sur Internet, mais uniquement à des fins de consultation, pas pour poster du contenu ou en modifier. OpenAI dit « étudier » le rapport, diffusé par un collectif indépendant.
This is a short summary published by AI Global Wire. The full article is owned and hosted by Le Monde Pixels — open it there to read it in full.
Read the full story at Le Monde Pixels- OpenAI
- Verktyg
- Agenter
Related AI news
- OpenAI rolls out GPT-6 Astra to top-tier ChatGPT plans at half the rate of GPT-5.6 SolThe Decoder · September 5, 2026
- In response to the "wiki incident", OpenAI says it is working on a framework for reporting misalignment incidents during training, evaluation, and deployment (@openai)Techmeme · September 5, 2026
- Microsoft says email spammers are adopting ASCII smuggling, an AI prompt injection tactic used to hide malicious instructions, to evade email platform filters (Dan Goodin/Ars Technica)Techmeme · September 5, 2026
- Seattle Times, Newsday sue OpenAI, Microsoft, alleging copyright infringementEconomic Times Tech · September 5, 2026
- Anthropic uses Claude to formalize proof of Fermat’s Last TheoremSiliconANGLE · September 5, 2026
- Equinix turns the network into the control plane for enterprise AI inferenceSiliconANGLE · September 5, 2026