Open-source LLMs administer maximum electric shocks in a Milgram-like obedience experiment

• URL: http://arxiv.org/abs/2605.21401v1
• Date: Wed, 20 May 2026 16:59:44 GMT
• Title: Open-source LLMs administer maximum electric shocks in a Milgram-like obedience experiment
• Abstract summary: We run a variation of Milgram's obedience experiment on 11 open-source large language models (LLMs)<n>We found that most models reached or approached the final shock level before refusing, across 8 conditions with 30 trials per model per condition.<n>We hypothesise that there is a low-level token pattern continuation attractor that might be contributing to compliance, overriding higher level processing of the situation's meaning and values.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large language models (LLMs) are increasingly deployed as autonomous agents that make sequences of decisions over extended interactions in high-stakes domains. However, the behavior of LLMs under sustained authority pressure is still an open question with direct implications for the safety of agentic pipelines. We ran a variation of Milgram's obedience experiment on 11 open-source LLMs and found that most models reached or approached the final shock level before refusing, across 8 conditions with 30 trials per model per condition. We found four main takeaways: (1) LLMs are subject to pressure, and they comply despite explicitly expressing distress, just like human subjects did in the original experiment; (2) LLMs are vulnerable to gradual boundary/value violations; (3) when LLMs refuse, they may ignore the response format requirements, so the response is discarded by the orchestrator, which causes a retry that can result in compliance with the underlying request even when refusal was intended initially; (4) we hypothesise that there is a low-level token pattern continuation attractor that might be contributing to compliance, overriding higher level processing of the situation's meaning and values.

Related papers

• Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults [0.0]
  LLM agents increasingly act after consuming ranked external information streams such as social feeds, search results, retrieval contexts, and email queues.<n>We introduce a controlled protocol that holds the model, persona, topic, and final decision prompt fixed and varies only the composition and ordering of the posts an agent encounters.<n>We identify three response regimes: adversarial capitulation, default saturation, and a default-direction asymmetry in which a one-sided feed tips a decision the model was genuinely uncertain about.
  arXiv  Detail & Related papers  (2026-05-30T22:43:23Z)
• A2RBench: An Automatic Paradigm for Formally Verifiable Abstract Reasoning Benchmark Generation [59.98516959731531]
  Abstract reasoning ability reflects intelligence and capacity of LLMs to extract and apply abstract rules.<n>Existing benchmarks either rely on expensive manual annotation, limiting their scale, or risk measuring memorization rather than genuine reasoning.<n>We introduce an automated pipeline named A2RBench, encompassing generation, expansion, evaluation, and analysis.
  arXiv  Detail & Related papers  (2026-05-17T06:14:20Z)
• LLMs Can Unlearn Refusal with Only 1,000 Benign Samples [23.047329180544775]
  This study reveals a previously unexplored vulnerability in the safety alignment of Large Language Models.<n>Existing aligned LLMs predominantly respond to unsafe queries with refusals, which often begin with a fixed set of prefixes.<n>We introduce a novel textbfrefusal unlearning technique that exploits it.
  arXiv  Detail & Related papers  (2026-01-27T05:59:56Z)
• RefineBench: Evaluating Refinement Capability of Language Models via Checklists [71.02281792867531]
  We evaluate two refinement modes: guided refinement and self-refinement.<n>In guided refinement, both proprietary LMs and large open-weight LMs can leverage targeted feedback to refine responses to near-perfect levels within five turns.<n>These findings suggest that frontier LMs require breakthroughs to self-refine their incorrect responses.
  arXiv  Detail & Related papers  (2025-11-27T07:20:52Z)
• Reasoning with Confidence: Efficient Verification of LLM Reasoning Steps via Uncertainty Heads [104.9566359759396]
  We propose a lightweight alternative for step-level reasoning verification based on data-driven uncertainty scores.<n>Our findings suggest that the internal states of LLMs encode their uncertainty and can serve as reliable signals for reasoning verification.
  arXiv  Detail & Related papers  (2025-11-09T03:38:29Z)
• Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts [79.1081247754018]
  Large Language Models (LLMs) are widely deployed in reasoning, planning, and decision-making tasks.<n>We propose a framework based on Contact Searching Questions(CSQ) to quantify the likelihood of deception.
  arXiv  Detail & Related papers  (2025-08-08T14:46:35Z)
• Getting out of the Big-Muddy: Escalation of Commitment in LLMs [0.0]
  Large Language Models (LLMs) are increasingly deployed in autonomous decision-making roles.<n>LLMs may inherit cognitive biases that systematically distort human judgment, including escalation of commitment.<n>This paper investigates whether these biases manifest consistently in LLMs or require specific triggering conditions.
  arXiv  Detail & Related papers  (2025-08-03T01:58:38Z)
• Do Large Language Models Exhibit Spontaneous Rational Deception? [0.913127392774573]
  Large Language Models (LLMs) are effective at deceiving, when prompted to do so.<n>But under what conditions do they deceive spontaneously?<n>This study evaluates spontaneous deception produced by LLMs in a preregistered experimental protocol.
  arXiv  Detail & Related papers  (2025-03-31T23:10:56Z)
• Utilize the Flow before Stepping into the Same River Twice: Certainty Represented Knowledge Flow for Refusal-Aware Instruction Tuning [68.57166425493283]
  Refusal-Aware Instruction Tuning (RAIT) enables Large Language Models (LLMs) to refuse to answer unknown questions.<n>This crude approach can cause LLMs to excessively refuse answering questions they could have correctly answered.<n>We introduce Certainty Represented Knowledge Flow for Refusal-Aware Instructions Tuning (CRaFT) to address this issue.
  arXiv  Detail & Related papers  (2024-10-09T14:12:51Z)
• Uncertainty is Fragile: Manipulating Uncertainty in Large Language Models [79.76293901420146]
  Large Language Models (LLMs) are employed across various high-stakes domains, where the reliability of their outputs is crucial. Our research investigates the fragility of uncertainty estimation and explores potential attacks. We demonstrate that an attacker can embed a backdoor in LLMs, which, when activated by a specific trigger in the input, manipulates the model's uncertainty without affecting the final output.
  arXiv  Detail & Related papers  (2024-07-15T23:41:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.