Gary Marcus: OpenAI's new technique could destroy chain-of-thought monitorability

GaryMarcus · x · 2026-09-02

The Information reported that OpenAI is experimenting with a technique that makes models reveal less of their "thinking", making them harder to monitor. Gary Marcus warns OpenAI is crossing an AI safety redline:

Related event: OpenAI Reportedly Testing Tech That Obscures Chain-of-Thought, Sparking AI Safety Debate(8 posts)→

Original post →

More from Models

Models channel →