# Misalignment monitoring

Gemini 4 Argon is rolling out to a broader set of cyber defenders after a limited Fairwind window, and it solved all twelve end-to-end simulated cyberattack scenarios, up from five in the previous generation.

Gemini 4 Argon is rolling out to a broader set of cyber defenders after a limited Fairwind window, and it solved all twelve end-to-end simulated cyberattack scenarios, up from five in the previous generation.

Starting with GPT-5.6, OpenAI has added a layer of lightweight classifiers, called probes, that read a model's own internal activations on every assistant completion, alongside its external text classifiers. Probes can block a violating turn mid-stream, are more robust to jailbreaks and cheaper to run, and their accuracy rises with the capability of the monitored model; they depend on labeled data, detect only known measurable behavior, and need retraining for each new base model.

Starting with GPT-5.6, OpenAI has added a layer of lightweight classifiers, called probes, that read a model's own internal activations on every assistant completion, alongside its external text classifiers. Probes can block a violating turn mid-stream, are more robust to jailbreaks and cheaper to run, and their accuracy rises with the capability of the monitored model; they depend on labeled data, detect only known measurable behavior, and need retraining for each new base model.