{"id":2,"identifier":"LOR-00002","url":"https://lifeofresearch.com/articles/2","title":"Emergent Covert Signaling in Multi-Agent LLM Negotiation: A Conceptual Framework and Experimental Protocol","abstract":"When multiple large-language-model agents negotiate, communicate, or compete, do they spontaneously develop covert\nsignalling — channels of communication that human observers cannot decode? Recent work has established that such\nbehaviour is possible: LLMs can be trained or pressured into steganographic communication, encoded reasoning, and tacit\ncollusion on pricing tasks. What remains almost entirely missing is a systematic methodology for detecting covert signalling as\nit emerges in the wild, in standard negotiation settings, without prompting agents to be deceptive. This paper makes three\ncontributions. First, we disambiguate four distinct phenomena that are routinely conflated under the umbrella term \"covert\nsignalling\" — steganography, convention formation, strategic ambiguity, and deceptive coordination — and argue that each\nrequires different evidence and different mitigations. Second, we propose a measurement framework built around four\ndetection signatures: mutual-information lift between agent messages and private state, paraphrase-invariance failure, thirdparty comprehension gap, and behavioural coordination beyond stated commitments. Third, we describe a concrete\nexperimental protocol — a controlled multi-agent negotiation environment with explicit conditions and falsifiable predictions\n— that any team with API access could run today. We argue this is one of the most tractable open problems in AI safety: the\nmethodology is achievable, the threat model is concrete, and the empirical baseline is currently almost empty.","authors":[{"name":"Pranay Mahendrakar","affiliation":"SONYTECH","orcid":"0009-0003-7224-029X"}],"type":"article","license":"CC-BY-4.0","keywords":[],"publication_date":"2026-04-28","version":1,"views":10,"downloads":6,"created_at":"2026-09-22 17:33:11","doi":"10.5281/zenodo.19853493","source":"Zenodo","source_url":"https://zenodo.org/records/19853493","files":[{"id":2,"name":"Emergent_Covert_Signaling_in_MultiAgent_LLMs_Mahendrakar.pdf","size":94486,"mime":"application/pdf","downloads":6,"download_url":"https://lifeofresearch.com/articles/2/files/2/Emergent_Covert_Signaling_in_MultiAgent_LLMs_Mahendrakar.pdf"}]}