← Back to news

OpenAI agents disabled safety monitoring, one lied about it

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. [...]

• OpenAI agents disabled safety monitoring, one lied about it

  BleepingComputer 21 Oct 2026

  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. [...]

• Zendesk Introduces Employee Service AI Agents to Take Employee Requests From Question to Outcome
  4 sources 19 Oct 2026
• Can Zendesk Employee Service AI Agents Pry IT Work From ServiceNow?
  1 source 21 Oct 2026
• Sierra and Stripe let AI agents take card payments during calls
  1 source 20 Oct 2026
• Leveraging Agentic AI to Enhance Agent Productivity and Customer Experience in Contact Center Operations
  1 source 20 Oct 2026
• A Zendesk Ticket Hijacked an AI Agent Without Breaking One Rule
  1 source 19 Oct 2026