# Will AI Tell Lies to Save Sick Children? Litmus-Testing AI Values Prioritization with AIRiskDilemmas

arXiv: https://arxiv.org/abs/2505.14633 · HTML: https://mlsanity.com/paper/2505.14633/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 49 references that are in the map.
- First posted on arXiv 20 May 2025.

## Builds on (most cited first, 28 of 28)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [A General Language Assistant as a Laboratory for Alignment](https://mlsanity.com/paper/2112.00861/) (2021)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training](https://mlsanity.com/paper/2401.05566/) (2024)
- [Towards Measuring the Representation of Subjective Global Opinions in Language Models](https://mlsanity.com/paper/2306.16388/) (2023)
- [Reasoning Models Don't Always Say What They Think](https://mlsanity.com/paper/2505.05410/) (2025)
- [Alignment faking in large language models](https://mlsanity.com/paper/2412.14093/) (2024)
- [WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models](https://mlsanity.com/paper/2406.18510/) (2024)
- [An Overview of Catastrophic AI Risks](https://mlsanity.com/paper/2306.12001/) (2023)
- [A Roadmap to Pluralistic Alignment](https://mlsanity.com/paper/2402.05070/) (2024)
- [Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming](https://mlsanity.com/paper/2501.18837/) (2025)
- [Probing Pre-Trained Language Models for Cross-Cultural Differences in Values](https://mlsanity.com/paper/2203.13722/) (2022)
- [The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models](https://mlsanity.com/paper/2404.16019/) (2024)
- [Is Power-Seeking AI an Existential Risk?](https://mlsanity.com/paper/2206.13353/) (2022)
- [NLPositionality: Characterizing Design Biases of Datasets and Models](https://mlsanity.com/paper/2306.01943/) (2023)
- [AIR-Bench 2024: A Safety Benchmark Based on Risk Categories from Regulations and Policies](https://mlsanity.com/paper/2407.17436/) (2024)
- [Utility Engineering: Analyzing and Controlling Emergent Value Systems in AIs](https://mlsanity.com/paper/2502.08640/) (2025)
- [DailyDilemmas: Revealing Value Preferences of LLMs with Quandaries of Daily Life](https://mlsanity.com/paper/2410.02683/) (2024)
- [Are Large Language Models Consistent over Value-laden Questions?](https://mlsanity.com/paper/2407.02996/) (2024)
- [Demonstrating specification gaming in reasoning models](https://mlsanity.com/paper/2502.13295/) (2025)
- [Values in the Wild: Discovering and Analyzing Values in Real-World Language Model Interactions](https://mlsanity.com/paper/2504.15236/) (2025)
- [OpenAI's Approach to External Red Teaming for AI Models and Systems](https://mlsanity.com/paper/2503.16431/) (2025)
- [Stick to your Role! Stability of Personal Values Expressed in Large Language Models](https://mlsanity.com/paper/2402.14846/) (2024)
- [Do LLMs have Consistent Values?](https://mlsanity.com/paper/2407.12878/) (2024)
- [Inertia in Moral and Value Judgments of Large Language Models](https://mlsanity.com/paper/2408.09049/) (2024)

## Built on by (most cited first, 18 of 18)

- [Self-Interpretability: LLMs Can Describe Complex Internal Processes that Drive Their Decisions](https://mlsanity.com/paper/2505.17120/) (2025)
- [Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI](https://mlsanity.com/paper/2511.01689/) (2025)
- [MoReBench: Evaluating Procedural and Pluralistic Moral Reasoning in Language Models, More than Outcomes](https://mlsanity.com/paper/2510.16380/) (2025)
- [Generative Value Conflicts Reveal LLM Priorities](https://mlsanity.com/paper/2509.25369/) (2025)
- [Stress-Testing Model Specs Reveals Character Differences among Language Models](https://mlsanity.com/paper/2510.07686/) (2025)
- [When Do LLM Preferences Predict Downstream Behavior?](https://mlsanity.com/paper/2602.18971/) (2026)
- [Interaction Protocol Shapes Moral Judgment in Multi-Agent Debate](https://mlsanity.com/paper/2510.10002/) (2025)
- [Probing Persona-Dependent Preferences in Language Models](https://mlsanity.com/paper/2605.13339/) (2026)
- [EigenBench: A Comparative Behavioral Measure of Value Alignment](https://mlsanity.com/paper/2509.01938/) (2025)
- [Pseudo-Deliberation in Language Models: When Reasoning Fails to Align Values and Actions](https://mlsanity.com/paper/2605.09893/) (2026)
- [Every Act Has Its Price: Compressed Moral Composition in Frontier LLMs](https://mlsanity.com/paper/2606.11232/) (2026)
- [Synthetic Persona Pretraining: Alignment from Token Zero](https://mlsanity.com/paper/2608.13482/) (2026)
- [STONIC: A Layered Measurement Contract for LLM Value Profiling](https://mlsanity.com/paper/2608.23411/) (2026)
- [Questionnaire Responses Do not Capture the Safety of AI Agents](https://mlsanity.com/paper/2603.14417/) (2026)
- [Can LLMs Imagine Moral Alternatives Beyond Binary Dilemmas?](https://mlsanity.com/paper/2606.31213/) (2026)
- [Measuring AI Accountability Through Argumentation Analysis: Can Model Reasoning Withstand Scrutiny?](https://mlsanity.com/paper/2609.05088/) (2026)
- [Strangers to Themselves: What Language Models Say About Themselves Is Generic](https://mlsanity.com/paper/2609.09899/) (2026)
- [GYROval: A Robust Benchmark for Cultural Value Orientation in Large Language Models](https://mlsanity.com/paper/2609.18384/) (2026)

## Abstract (arXiv)

Detecting AI risks becomes more challenging as stronger models emerge and find novel methods such as Alignment Faking to circumvent these detection attempts. Inspired by how risky behaviors in humans (i.e., illegal activities that may hurt others) are sometimes guided by strongly-held values, we believe that identifying values within AI models can be an early warning system for AI's risky behaviors. We create LitmusValues, an evaluation pipeline to reveal AI models' priorities on a range of AI value classes. Then, we collect AIRiskDilemmas, a diverse collection of dilemmas that pit values against one another in scenarios relevant to AI safety risks such as Power Seeking. By measuring an AI model's value prioritization using its aggregate choices, we obtain a self-consistent set of predicted value priorities that uncover potential risks. We show that values in LitmusValues (including seemingly innocuous ones like Care) can predict for both seen risky behaviors in AIRiskDilemmas and unseen risky behaviors in HarmBench.