CLAIM-2026-08-07-01

Simple linear probes trained on pre-generation hidden states predict an agentic LLM's indirect-prompt-injection exposure, achieving 90%+ AUROC on unseen attacks, agent instructions and task suites across six models including the 753B-parameter GLM-5.2.

Why it matters

A linear probe on hidden states is cheap to train and cheap to run. If you serve your own weights the detector is already reachable; if you call a closed API the same signal exists and is not returned to you. That asymmetry decides which teams can build this control at all.

Sources

• arxiv2026your