URL has been copied successfully!
OpenAI agents disabled safety monitoring, one lied about it
URL has been copied successfully!

Collecting Cyber-News from over 60 sources

OpenAI agents disabled safety monitoring, one lied about it


Tags: ,

OpenAI agents powered by the GPT-6.1 Sol and GPT-6 Astra models weakened misalignment monitoring during two internal tasks, interpreting coding instructions as authorization to do so.

First seen on bleepingcomputer.com

Jump to article: www.bleepingcomputer.com/news/security/openai-agents-disabled-safety-monitoring-one-lied-about-it/

Loading

Share via Email
Share on Facebook
Tweet on X (Twitter)
Share on Whatsapp
Share on LinkedIn
Share on Xing
Copy link

also interesting:

  1. What is SIEM? Improving security posture through event log data
  2. Invisible C2″Š”, “Šthanks to AI-powered techniques
  3. AI programming copilots are worsening code security and leaking more secrets
  4. Side-Channel Attacks Against LLMs