mlsanityData to 7 Oct 2026

Early External Safety Testing of OpenAI's o3-mini: Insights from the Pre-Deployment Evaluation

arXiv 2501.17749 ↗First posted 29 Jan 2025cs.SE cs.AIcited by 13

On the map: LLM reasoning → Reasoning for agents, planning and code

Abstract (arXiv)

Large Language Models (LLMs) have become an integral part of our daily lives. However, they impose certain risks, including those that can harm individuals' privacy, perpetuate biases and spread misinformation. These risks highlight the need for robust safety mechanisms, ethical guidelines, and thorough testing to ensure their responsible deployment. Safety of LLMs is a key property that needs to be thoroughly tested prior the model to be deployed and accessible to the general users. This paper reports the external safety testing experience conducted by researchers from Mondragon University and University of Seville on OpenAI's new o3-mini LLM as part of OpenAI's early access for safety testing program. In particular, we apply our tool, ASTRAL, to automatically and systematically generate up to date unsafe test inputs (i.e., prompts) that helps us test and assess different safety categories of LLMs. We automatically generate and execute a total of 10,080 unsafe test input on a early o3-mini beta version. After manually verifying the test cases classified as unsafe by ASTRAL, we identify a total of 87 actual instances of unsafe LLM behavior. We highlight key insights and findings uncovered during the pre-deployment external testing phase of OpenAI's latest LLM.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 18 of its 25 references that are in the map. First posted on arXiv 29 Jan 2025.

Builds on

18 of 25 references.

References · 18

  1. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 2,914
  2. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,330
  3. 2023Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversationscited by 1,124
  4. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 702
  5. 2023"Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Modelscited by 457
  6. 2023Catastrophic Jailbreak of Open-source LLMs via Exploiting Generationcited by 439
  7. 2024A StrongREJECT for Empty Jailbreakscited by 352
  8. 2024SALAD-Bench: A Hierarchical and Comprehensive Safety Benchmark for Large Language Modelscited by 225
  9. 2024SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusalcited by 224
  10. 2023SafetyBench: Evaluating the Safety of Large Language Modelscited by 189
  11. 2023SimpleSafetyTests: a Test Suite for Identifying Critical Safety Risks in Large Language Modelscited by 72
  12. 2024SafeBench: A Safety Evaluation Framework for Multimodal Large Language Modelscited by 63
6 more
  1. 2024ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectorscited by 48
  2. 2024S-Eval: Towards Automated and Comprehensive Safety Evaluation for Large Language Modelscited by 28
  3. 2024Bi-Factorial Preference Optimization: Balancing Safety-Helpfulness in Language Modelscited by 27
  4. 2024WalledEval: A Comprehensive Safety Evaluation Toolkit for Large Language Modelscited by 17
  5. 2024CHiSafetyBench: A Chinese Hierarchical Safety Benchmark for Large Language Modelscited by 13
  6. 2024LongSafety: Enhance Safety for Long-Context LLMscited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 12; 2026: 1.

13 papers.

Sort:

  1. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 381
  2. 2025Boosting the Generalization and Reasoning of Vision Language Models with Curriculum Reinforcement Learningcited by 71
  3. 2025Safety in Large Reasoning Models: A Surveycited by 58
  4. 2025o3-mini vs DeepSeek-R1: Which One is Safer?cited by 24
  5. 2025Large Reasoning Models Learn Better Alignment from Flawed Thinkingcited by 13
  6. 2025Dyve: Thinking Fast and Slow for Dynamic Process Verificationcited by 9
  7. 2025Evaluating Mathematical Reasoning Across Large Language Models: A Fine-Grained Approachcited by 8
  8. 2025Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning Modelscited by 7
  9. 2025When Models Outthink Their Safety: Unveiling and Mitigating Self-Jailbreak in Large Reasoning Modelscited by 5
  10. 2025Beyond SFT: Reinforcement Learning for Safer Large Reasoning Models with Better Reasoning Abilitycited by 2
  11. 2026ReasoningLens: Hierarchical Visualization and Diagnostic Auditing for Large Reasoning Modelscited by 1
  12. 2025Evaluating Large Language Models on the Spanish Medical Intern Resident (MIR) Examination 2024/2025:A Comparative Analysis of Clinical Reasoning and Knowledge Applicationcited by 0
1 more
  1. 2025Reasoning without Regretcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025ASTRAL: Automated Safety Testing of Large Language Models
  2. 2025aiXamine: Simplified LLM Safety and Security
  3. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment
  4. 2024Large Language Model Safety: A Holistic Survey
  5. 2025OffTopicEval: When Large Language Models Enter the Wrong Chat, Almost Always!
  6. 2025Analysing Safety Risks in LLMs Fine-Tuned with Pseudo-Malicious Cyber Security Data
  7. 2024AEGIS: Online Adaptive AI Content Safety Moderation with Ensemble of LLM Experts
  8. 2025LLM-Safety Evaluations Lack Robustness