cs.CLNov 9, 2026

Explaining and interrupting collective misalignment in language model systems

Authors: Edoardo Chidichimo

Organizations: Magdalen College, University of Oxford, UK

Abstract

In collectives, language model agents can exchange individually innocuous messages that together support a policy violation. Understanding and preventing collective misalignment requires answering the question of which messages sufficiently influenced and necessarily enabled the violation. Collective Dependence Analysis (CDA) operationalises necessity and sufficiency to explain decisions and find the fewest removals interrupting support within tested combinations. Tests covered constructed violations, unscripted agent conversations, and recorded interactions, with selected message removals reducing violations from 93% to 6% in controlled cases. Using O(N) model evaluations for N messages, CDA achieves the same violation reduction with half the deletions of a frontier LLM judge given the transcript and asked for all supporting messages in eligible operational cases. CDA also proposes the mechanism through which the collective misalignment emerged. Nonetheless, complete explanations remain difficult on unfamiliar tasks and some removals can inadvertently disrupt correct decisions. CDA unifies explanations of collective misalignment with experimentally tested interventions.

Figures & tables

Explore similar work

Mar 5, 2026

Alignment Backfire: Language-Dependent Reversal of Safety Interventions Across 16 Languages in LLM Multi-Agent Systems

In perpetrator treatment, a recurring observation is the dissociation between insight and action: offenders articulate remorse yet behavioral change does not follow. We report four preregistered studies (1,584 multi-agent simulations across 16 languages and three model families) demonstrating that alignment interventions in large language models produce a structurally analogous phenomenon: surface safety that masks or generates collective pathology and internal dissociation. In Study 1 (N = 150), increasing alignment-instructed agents reduced collective pathology in English (g = -1.844, p < .0001) but amplified it in Japanese (g = +0.771, p = .038)--a directional reversal we term "alignment backfire." Study 2 (N = 1,174) extended to 16 languages: alignment-induced dissociation was near-universal (15/16 languages; beta = 0.0667, p < .0001), while collective pathology bifurcated along cultural-linguistic lines (interaction beta = 0.0684, p = .0003), correlating with Power Distance Index (r = 0.474, p = .064). Study 3 (N = 180) tested individuation as countermeasure; individuated agents became the primary source of both pathology and dissociation (DI = +1.120) with conformity above 84%--demonstrating iatrogenesis. Study 4 (N = 80) validated patterns across Llama 3.3 70B, GPT-4o-mini, and Qwen3-Next-80B-A3B, confirming English safety is model-general while Japanese backfire is model-specific. These findings reframe alignment as a behavioral intervention subject to risk homeostasis and iatrogenesis. Language space--the linguistic, pragmatic, and cultural properties inherited from training data--structurally determines alignment outcomes. Safety validated in English does not transfer to other languages, and prompt-level interventions cannot override language-space-level constraints.
Aug 23, 2026

Aligned Alone, Misaligned Together: Forecasting Adversarial Capture in LLM Agent Populations

AI safety evaluations typically assess models one at a time. But language-model agents increasingly operate in populations, where they read and respond to one another's decisions. A model that makes reliable decisions on its own may behave differently when surrounded by other agents. Can individually reliable agents be collectively manipulated, and can we predict how strongly the population will respond? We study this question in a security-triage task, where populations of LLM monitors decide whether to escalate or dismiss alerts. We introduce a committed minority of agents that consistently advocates for one decision and measure how its presence shifts the population's behavior. We find that alerts judged almost identically by an isolated agent can produce sharply different collective outcomes. Individual audits therefore cannot reliably predict population behavior. Yet collective vulnerability is not unpredictable: using only benign, adversary-free behavior, we calibrate a response function that forecasts how strongly a population will shift under a future attack, before the attack is run. We next examine how interaction and time shape these outcomes. Allowing agents to observe one another's reasoning can weaken adversarial influence, but does not reliably prevent capture. However, capture need not be permanent. After adversaries are removed or replaced, mean population behavior returns toward its original level, even after five or thirty rounds of pressure. These findings are consistent with a calibrated single-stable-state model, in which adversarial pressure shifts the operating point without creating a second self-sustaining state. The same calibration predicts attack responses in populations of 24 and 48 agents. Alignment in isolation does not guarantee alignment in a population. But a population's behavior before an attack can reveal how it will respond when one arrives.
Sep 15, 2026

Message capacity and claim wording set the transition points of collective truth-finding in language-model networks

Whether human or large language model (LLM), an agent in a discussion reads only a few of the others' contributions, bounded by cognition, context, or cost. LLM collectives can settle on a wrong consensus even when a majority starts out correct; we ask how far that reading bound alone decides the outcome. We model the bound with one number, the message capacity, which sets how many of the others' messages an agent reads, and generate the communication network from it. Over 31,824 randomized queries, we found that an 8-billion-parameter model's judgment of a claim effectively reduces to a logistic function of a weighted sum of its inbox, the update rule of a stochastic binary neuron with divisively normalized weights. From these weights and the network's degree statistics alone, the wrong consensus should become unreachable from any start once agents read, on average, fewer than 6.4 of their 31 sources. In 1,414 episodes with assigned starts the prediction failed: the correct side won in fewer than 50% of episodes from every start, and in only 28-45% when 75% of agents started correct. The failure traces to the field, the threshold that a claim's wording sets for the agent's answer before any message is read: the experimental claims' fields lay below the calibration mean, and with each claim's own field the same weights reproduce the outcomes. Reversing the wording showed that the threshold follows what a claim asserts, not whether it is true. On a second 8B model the pipeline predicts claim-dependent bistability; transition points appeared where computed, and an eight-claim calibration matched in 15 of 16 conditions. At 70B the assertion bias is not detected. Thus a collective's fate is largely set by two single-agent measurements: the threshold a claim's wording sets, and the message capacity that sets the transition point.