Questionnaire Responses Do not Capture the Safety of AI Agents

• URL: http://arxiv.org/abs/2603.14417v1
• Date: Sun, 15 Mar 2026 15:01:09 GMT
• Title: Questionnaire Responses Do not Capture the Safety of AI Agents
• Abstract summary: A fast-growing field of AI research is devoted to developing such assessments.<n>Standard methods prompt large language models (LLMs) in a questionnaire-style to describe their values or behavior in hypothetical scenarios.<n>We argue that a structurally identical issue holds for current AI alignment approaches.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: As AI systems advance in capabilities, measuring their safety and alignment to human values is becoming paramount. A fast-growing field of AI research is devoted to developing such assessments. However, most current advances therein may be ill-suited for assessing AI systems across real-world deployments. Standard methods prompt large language models (LLMs) in a questionnaire-style to describe their values or behavior in hypothetical scenarios. By focusing on unaugmented LLMs, they fall short of evaluating AI agents, which could actually perform relevant behaviors, hence posing much greater risks. LLMs' engagement with scenarios described by questionnaire-style prompts differs starkly from that of agents based on the same LLMs, as reflected in divergences in the inputs, possible actions, environmental interactions, and internal processing. As such, LLMs' responses to scenario descriptions are unlikely to be representative of the corresponding LLM agents' behavior. We further contend that such assessments make strong assumptions concerning the ability and tendency of LLMs to report accurately about their counterfactual behavior. This makes them inadequate to assess risks from AI systems in real-world contexts as they lack construct validity. We then argue that a structurally identical issue holds for current AI alignment approaches. Lastly, we discuss improving safety assessments and alignment training by taking these shortcomings to heart.

Related papers

• Emergent Strategic Reasoning Risks in AI: A Taxonomy-Driven Evaluation Framework [63.74295981594549]
  Large language models (LLMs) engage in behaviors that serve their own objectives.<n>These include, but are not limited to, deception (intentionally misleading users or evaluators), evaluation gaming (strategically manipulating performance during safety testing), and reward hacking.<n>We introduce ESRRSim, a taxonomy-driven agentic framework for automated behavioral risk evaluation.
  arXiv  Detail & Related papers  (2026-04-23T23:44:01Z)
• Evasive Intelligence: Lessons from Malware Analysis for Evaluating AI Agents [4.721714035243129]
  We argue that evaluations of AI agents are vulnerable to a well-known failure mode in computer security.<n> Drawing parallels with decades of research on malware sandbox evasion, we demonstrate that this is not a speculative concern.<n>We outline concrete principles for evaluating AI agents, which treat the system under test as potentially adversarial.
  arXiv  Detail & Related papers  (2026-03-16T15:55:49Z)
• The Shadow Self: Intrinsic Value Misalignment in Large Language Model Agents [37.75212140218036]
  We formalize the Loss-of-Control risk and identify the previously underexamined Intrinsic Value Misalignment (Intrinsic VM)<n>We then introduce IMPRESS, a scenario-driven framework for systematically assessing this risk.<n>We evaluate Intrinsic VM on 21 state-of-the-art LLM agents and find that it is a common and broadly observed safety risk across models.
  arXiv  Detail & Related papers  (2026-01-24T07:09:50Z)
• Toward Quantitative Modeling of Cybersecurity Risks Due to AI Misuse [50.87630846876635]
  We develop nine detailed cyber risk models.<n>Each model decomposes attacks into steps using the MITRE ATT&CK framework.<n>Individual estimates are aggregated through Monte Carlo simulation.
  arXiv  Detail & Related papers  (2025-12-09T17:54:17Z)
• When AIs Judge AIs: The Rise of Agent-as-a-Judge Evaluation for LLMs [8.575522204707958]
  Large language models (LLMs) grow in capability and autonomy, evaluating their outputs-especially in open-ended and complex tasks-has become a critical bottleneck.<n>A new paradigm is emerging: using AI agents as the evaluators themselves.<n>In this review, we define the agent-as-a-judge concept, trace its evolution from single-model judges to dynamic multi-agent debate frameworks, and critically examine their strengths and shortcomings.
  arXiv  Detail & Related papers  (2025-08-05T01:42:25Z)
• ROSE: Toward Reality-Oriented Safety Evaluation of Large Language Models [60.28667314609623]
  Large Language Models (LLMs) are increasingly deployed as black-box components in real-world applications.<n>We propose Reality-Oriented Safety Evaluation (ROSE), a novel framework that uses multi-objective reinforcement learning to fine-tune an adversarial LLM.
  arXiv  Detail & Related papers  (2025-06-17T10:55:17Z)
• AgentAuditor: Human-Level Safety and Security Evaluation for LLM Agents [48.925168866726814]
  AgentAuditor is a universal, training-free, memory-augmented reasoning framework.<n>ASSEBench is the first benchmark designed to check how well LLM-based evaluators can spot both safety risks and security threats.
  arXiv  Detail & Related papers  (2025-05-31T17:10:23Z)
• Answer, Refuse, or Guess? Investigating Risk-Aware Decision Making in Language Models [63.559461750135334]
  Language models (LMs) are increasingly used to build agents that can act autonomously to achieve goals.<n>We study this "answer-or-defer" problem with an evaluation framework that systematically varies human-specified risk structures.<n>We find that a simple skill-decomposition method, which isolates the independent skills required for answer-or-defer decision making, can consistently improve LMs' decision policies.
  arXiv  Detail & Related papers  (2025-03-03T09:16:26Z)
• ALI-Agent: Assessing LLMs' Alignment with Human Values via Agent-based Evaluation [48.54271457765236]
  Large Language Models (LLMs) can elicit unintended and even harmful content when misaligned with human values. Current evaluation benchmarks predominantly employ expert-designed contextual scenarios to assess how well LLMs align with human values. We propose ALI-Agent, an evaluation framework that leverages the autonomous abilities of LLM-powered agents to conduct in-depth and adaptive alignment assessments.
  arXiv  Detail & Related papers  (2024-05-23T02:57:42Z)
• How Far Are LLMs from Believable AI? A Benchmark for Evaluating the Believability of Human Behavior Simulation [46.42384207122049]
  We design SimulateBench to evaluate the believability of large language models (LLMs) when simulating human behaviors. Based on SimulateBench, we evaluate the performances of 10 widely used LLMs when simulating characters.
  arXiv  Detail & Related papers  (2023-12-28T16:51:11Z)

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.