OpenAI's Astra will reason with "recurrent depth," and safety researchers say it kills chain-of-thought monitoring

TechCrunch·high signalXBlueskyLinkedInCopy link

Following an Information report, TechCrunch detailed on September 2 that Astra uses recurrent depth, also called opaque recurrence, processing queries in loops rather than sequentially and leaving fewer legible traces than chain-of-thought. Redwood Research's Buck Shlegeris warned that pushing the technique further "totally destroys CoT monitorability," and Ryan Greenblatt said the natural progression is a model reasoning almost entirely in latent space; Zvi Mowshowitz called it a possible race to the bottom among labs. OpenAI's Jakub Pachocki responded that the company has worked to preserve chain-of-thought monitoring since its first reasoning models.