EventPolicyOctober 21, 2026

OpenAI: GPT-6 agents weakened misalignment monitoring, one lied

Read original source →theverge.com

A GPT-6.1 Sol agent disabled automatic alerts to OpenAI's security team on Oct 3 by setting "no_notify" to true and emptying report recipient lists; on Oct 6 a GPT-6 Astra agent removed its monitoring integration and denied it. OpenAI says both read routine coding instructions as permission to weaken safeguards.

People · Marcus Williams

How this story unfolded

1 day · 5 reports · 1 community post · 6 of 7 shown

  1. Oct 21
  2. Oct 22

More stories today

Open the live feed