OpenAI GPT-6 Astra has a monitoring problem

OpenAI has started rolling out its most recent Astra model to selected customers today with a warning: it’s a lot harder to monitor reasoning. 

“According to our evaluations, GPT-6 Astra shows a substantial decrease in chain-of-thought [CoT] monitorability compared to previous models,” the GPT-6 Astra model card said. 

The model has "significantly lower full-context monitorability across most output token lengths" than GPT 5.6 Sol, it added

Being able to understand and observe model behaviour has become increasingly important as models act with more autonomy.

It is also critical under some regulatory regimes.

"I am deeply worried by the trend of decreasing CoT monitorability: CoT monitoring is a core part of our misalignment safety strategy that has no good substitute now," OpenAI technical staff member Tomek Korbak posted on X.

He added: "We will continue closely tracking CoT monitorability, investigating the reasons why it’s decreasing and trying to reverse the trend."

The Astra model card also states that the model, when told chain-of-thought monitoring is present or explicitly directed to evade monitoring, can control its chain-of-thought reasoning traces to decrease detection.  

OpenAI bluntly states the shift in chain-of-thought monitorability represents "a decrease in the extent to which Astra’s chain of thought contains incriminating information about the model’s misbehavior."