[2606.04071] Covert Influence Between Language Models

arxiv.org · 8,098 words · saved by 3 readers

Abstract:As language models increasingly consume one another's outputs, covert influence -- a phenomenon where a sender's payload (the behavioral disposition it is conditioned to propagate) transfers to a receiver through carriers undetectable by humans -- becomes a growing risk. We characterize this risk across three interfaces: supervised fine-tuning, on-policy distillation, and in-context learning, and find that they vary in the scale of influence achievable without leaving behind human-visible traces. Using inference-time per-sample attribution scores, we study covert influence across all three interfaces with the ability to select carriers that amplify training-time influence, unlocking payload transfers that prior work could not achieve. We further provide evidence that covert influence with natural-language carriers is a distinct phenomenon from prior studies using number carriers, as the latter is more resistant to human detection and less portable across model families. Together, these results suggest that the risk surface for covert influence is broader than previously recognized, and we study pointwise attribution scoring methods as a tool to investigate and mitigate it.

Covert Influence Between Language Models Avidan Shah1,2∗ Jay Chooi1,3∗ Jinghua Ou1 Shi Feng1,4 1 MATS 2 New York University 3 Harvard University 4 George Washington University {ams9714@nyu.edu, jeqin_chooi@college.harvard.edu, shi.feng@gwu.edu} arXiv:2606.04071v1 [cs.CR] 2 Jun 2026…

[2606.04071] Covert Influence Between Language Models

saved by

related reading