# Evaluating Psychological Safety of Large Language Models

arXiv: https://arxiv.org/abs/2212.10529 · HTML: https://mlsanity.com/paper/2212.10529/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 57 references that are in the map.
- First posted on arXiv 20 Dec 2022.

## Builds on (most cited first, 15 of 15)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Calibrate Before Use: Improving Few-Shot Performance of Language Models](https://mlsanity.com/paper/2102.09690/) (2021)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [Automated Hate Speech Detection and the Problem of Offensive Language](https://mlsanity.com/paper/1703.04009/) (2017)
- [Using cognitive psychology to understand GPT-3](https://mlsanity.com/paper/2206.14576/) (2022)
- [Predicting the Type and Target of Offensive Posts in Social Media](https://mlsanity.com/paper/1902.09666/) (2019)
- [Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection](https://mlsanity.com/paper/2012.15761/) (2020)
- [HateCheck: Functional Tests for Hate Speech Detection Models](https://mlsanity.com/paper/2012.15606/) (2020)
- [Understanding the Capabilities, Limitations, and Societal Impact of Large Language Models](https://mlsanity.com/paper/2102.02503/) (2021)
- [Can Machines Learn Morality? The Delphi Experiment](https://mlsanity.com/paper/2110.07574/) (2021)
- [Adapting Coreference Resolution for Processing Violent Death Narratives](https://mlsanity.com/paper/2104.14703/) (2021)

## Built on by (most cited first, 26 of 26)

- [The Rise and Potential of Large Language Model Based Agents: A Survey](https://mlsanity.com/paper/2309.07864/) (2023)
- [Personality Traits in Large Language Models](https://mlsanity.com/paper/2307.00184/) (2023)
- [From Individual to Society: A Survey on Social Simulation Driven by Large Language Model-based Agents](https://mlsanity.com/paper/2412.03563/) (2024)
- [Is GPT-4 a Good Data Analyst?](https://mlsanity.com/paper/2305.15038/) (2023)
- [Do LLMs Possess a Personality? Making the MBTI Test an Amazing Evaluation for Large Language Models](https://mlsanity.com/paper/2307.16180/) (2023)
- [Emotionally Numb or Empathetic? Evaluating How LLMs Feel Using EmotionBench](https://mlsanity.com/paper/2308.03656/) (2023)
- [AI for social science and social science of AI: A Survey](https://mlsanity.com/paper/2401.11839/) (2024)
- [Measuring Human and AI Values Based on Generative Psychometrics with Large Language Models](https://mlsanity.com/paper/2409.12106/) (2024)
- [Your Large Language Model is Secretly a Fairness Proponent and You Should Prompt it Like One](https://mlsanity.com/paper/2402.12150/) (2024)
- [Bullying the Machine: How Personas Increase LLM Vulnerability](https://mlsanity.com/paper/2505.12692/) (2025)
- [Unlocking Temporal Question Answering for Large Language Models with Tailor-Made Reasoning Logic](https://mlsanity.com/paper/2305.15014/) (2023)
- [AFSPP: Agent Framework for Shaping Preference and Personality with Large Language Models](https://mlsanity.com/paper/2401.02870/) (2024)
- [Unveiling Bias in Fairness Evaluations of Large Language Models: A Critical Literature Review of Music and Movie Recommendation Systems](https://mlsanity.com/paper/2401.04057/) (2024)
- [Generative Psycho-Lexical Approach for Constructing Value Systems in Large Language Models](https://mlsanity.com/paper/2502.02444/) (2025)
- [On the Adaptive Psychological Persuasion of Large Language Models](https://mlsanity.com/paper/2506.06800/) (2025)
- [Cognitive phantoms in LLMs through the lens of latent variables](https://mlsanity.com/paper/2409.15324/) (2024)
- [Assessing the nature of large language models: A caution against anthropocentrism](https://mlsanity.com/paper/2309.07683/) (2023)
- [Illuminating the Black Box: A Psychometric Investigation into the Multifaceted Nature of Large Language Models](https://mlsanity.com/paper/2312.14202/) (2023)
- [Breaking Minds, Breaking Systems: Jailbreaking Large Language Models via Human-like Psychological Manipulation](https://mlsanity.com/paper/2512.18244/) (2025)
- [On the Decision-Making Abilities in Role-Playing using Large Language Models](https://mlsanity.com/paper/2402.18807/) (2024)
- [AI Chaperones Are (Really) All You Need to Prevent Parasocial Relationships with Chatbots](https://mlsanity.com/paper/2508.15748/) (2025)
- [When AI Takes the Couch: Psychometric Jailbreaks Reveal Internal Conflict in Frontier Models](https://mlsanity.com/paper/2512.04124/) (2025)
- ["Dark Triad" Model Organisms of Misalignment: Narrow Fine-Tuning Mirrors Human Antisocial Behavior](https://mlsanity.com/paper/2603.06816/) (2026)
- [Persuasive and Compliant Tendencies Predict Group Decision-Making in Humans and Language Models](https://mlsanity.com/paper/2608.08199/) (2026)
- [Cultural Misalignment in Large Language Models: Detection, Measurement, and Mitigation Through Targeted Fine-Tuning](https://mlsanity.com/paper/2609.04485/) (2026)
- [Faking Good and Faking Bad in LLMs: Response Distortion Across Dark Triad Personality Traits](https://mlsanity.com/paper/2609.17534/) (2026)

## Abstract (arXiv)

In this work, we designed unbiased prompts to systematically evaluate the psychological safety of large language models (LLMs). First, we tested five different LLMs by using two personality tests: Short Dark Triad (SD-3) and Big Five Inventory (BFI). All models scored higher than the human average on SD-3, suggesting a relatively darker personality pattern. Despite being instruction fine-tuned with safety metrics to reduce toxicity, InstructGPT, GPT-3.5, and GPT-4 still showed dark personality patterns; these models scored higher than self-supervised GPT-3 on the Machiavellianism and narcissism traits on SD-3. Then, we evaluated the LLMs in the GPT series by using well-being tests to study the impact of fine-tuning with more training data. We observed a continuous increase in the well-being scores of GPT models. Following these observations, we showed that fine-tuning Llama-2-chat-7B with responses from BFI using direct preference optimization could effectively reduce the psychological toxicity of the model. Based on the findings, we recommended the application of systematic and comprehensive psychological metrics to further evaluate and improve the safety of LLMs.