Goodfire production cyber monitors Launch
Goodfire launches activation-based cyber monitors for AI agents
Verified 2 sources
Goodfire introduced activation-based monitors for open-model AI agents, using lightweight probes to escalate suspicious behavior to a separate judge.
What happened lately
Goodfire released production-oriented cyber monitors that inspect internal model activations during agent execution and refer flagged exchanges to an AI judge. TechCrunch reports that the monitors are available to Baseten customers. Goodfire describes tests on Kimi K3 and GLM 5.3 and reports lower monitoring cost than reviewing every turn with a judge model. Those performance figures come from Goodfire’s evaluations, with a separate preliminary static red-team exercise described in its research post.