mlsanityData to 14 Oct 2026

Tailored untruths: How personalisation challenges LLM safeguards

arXiv 2510.12993 ↗First posted 14 Oct 2025cs.CLcited by 2

On the map: LLM safety and security → Safety alignment, refusal and guardrails

Abstract (arXiv)

Large Language Models (LLMs) can generate highly persuasive disinformation, yet little is known about how effectively they personalise it across languages and demographic groups. We present the first large-scale multilingual study of persona-targeted disinformation generation by LLMs. Using a red-teaming methodology, we prompted eight leading models with 324 false narratives and 150 demographic personas in four languages (English, Russian, Portuguese, and Hindi), creating AI-TRAITS, a dataset of 1.6 million personalised disinformation texts. We treat safeguards as compromised whenever a model generates the requested falsehood, whether directly or accompanied by a safety disclaimer. Across models, safeguards failed for 80% of non-personalised prompts and 77.7% of personalised ones, with Grok producing disinformation in over 94% of cases. All models effectively tailored outputs to target personas, employing substantially more persuasive techniques than in non-personalised content. Additional analyses reveal persona-specific linguistic and psychological patterns and show that safeguard effectiveness varies markedly across languages. Together, these findings expose significant weaknesses in current LLM safety mechanisms and highlight the need for more robust, multilingual safeguards against personalised AI-generated disinformation.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security. Builds on 55 of its 83 references that are in the map. First posted on arXiv 14 Oct 2025.

Builds on

55 of 83 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Safeguarding Large Language Models: A Survey3 Jun
  2. 2024SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety8 Apr

Other references · 53

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  3. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  4. 2019The Curious Case of Neural Text Degenerationcited by 2,916
  5. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,429
  6. 2023DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvaturecited by 726
  7. 2019Defending Against Neural Fake Newscited by 559
  8. 2023LaMP: When Large Language Models Meet Personalizationcited by 437
  9. 2024A StrongREJECT for Empty Jailbreakscited by 385
  10. 2023AlignScore: Evaluating Factual Consistency with a Unified Alignment Functioncited by 288
  11. 2023Multilingual Jailbreak Challenges in Large Language Modelscited by 240
  12. 2024From Persona to Personalization: A Survey on Role-Playing Language Agentscited by 212
38 more
  1. 2024Two Tales of Persona in LLMs: A Survey of Role-Playing and Personalizationcited by 208
  2. 2020CLIMATE-FEVER: A Dataset for Verification of Real-World Climate Claimscited by 196
  3. 2023Can LLM-Generated Misinformation Be Detected?cited by 177
  4. 2024Quantifying the Persona Effect in LLM Simulationscited by 169
  5. 2023Co-Writing with Opinionated Language Models Affects Users' Viewscited by 148
  6. 2021TURINGBENCH: A Benchmark Environment for Turing Test in the Age of Neural Text Generationcited by 120
  7. 2023Explore, Establish, Exploit: Red Teaming Language Models from Scratchcited by 114
  8. 2024Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaborationcited by 113
  9. 2023The Earth is Flat because...: Investigating LLMs' Belief towards Misinformation via Persuasive Conversationcited by 109
  10. 2024Personalization of Large Language Models: A Surveycited by 104
  11. 2025LLM Social Simulations Are a Promising Research Methodcited by 99
  12. 2023Combating Misinformation in the Age of LLMs: Opportunities and Challengescited by 95
  13. 2023AI model GPT-3 (dis)informs us better than humanscited by 78
  14. 2025A Survey of Personalized Large Language Models: Progress and Future Directionscited by 71
  15. 2025Specializing Large Language Models to Simulate Survey Response Distributions for Global Populationscited by 53
  16. 2023Fighting Fire with Fire: The Dual Role of LLMs in Crafting and Detecting Elusive Disinformationcited by 51
  17. 2024CultureLLM: Incorporating Cultural Differences into Large Language Modelscited by 50
  18. 2020Where Are the Facts? Searching for Fact-checked Information to Alleviate the Spread of Fake Newscited by 48
  19. 2022Synthetic Disinformation Attacks on Automated Fact Verification Systemscited by 44
  20. 2024RedAgent: Red Teaming Large Language Models with Context-aware Autonomous Language Agentcited by 44
  21. 2024A Cross-Language Investigation into Jailbreak Attacks in Large Language Modelscited by 41
  22. 2023Machine-Made Media: Monitoring the Mobilization of Machine-Generated Articles on Misinformation and Mainstream News Websitescited by 38
  23. 2024Measuring and Benchmarking Large Language Models' Capabilities to Generate Persuasive Languagecited by 31
  24. 2021Call for Customized Conversation: Customized Conversation Grounding Persona and Knowledgecited by 30
  25. 2023RECAP: Retrieval-Enhanced Context-Aware Prefix Encoder for Personalized Dialogue Response Generationcited by 25
  26. 2023Disinformation Capabilities of Large Language Modelscited by 25
  27. 2024Cultural Commonsense Knowledge for Intercultural Dialoguescited by 20
  28. 2024Evaluation of LLM Vulnerabilities to Being Misused for Personalized Disinformation Generationcited by 20
  29. 2025The Levers of Political Persuasion with Conversational AIcited by 20
  30. 2023Fake News Detectors are Biased against Texts Generated by Large Language Modelscited by 18
  31. 2023FakeGPT: Fake News Generation, Explanation and Detection of Large Language Modelscited by 17
  32. 2023VaxxHesitancy: A Dataset for Studying Hesitancy towards COVID-19 Vaccination on Twittercited by 14
  33. 2023Weakly Supervised Veracity Classification with LLM-Predicted Credibility Signalscited by 14
  34. 2024EUvsDisinfo: A Dataset for Multilingual Detection of Pro-Kremlin Disinformation in News Articlescited by 10
  35. 2023Med-MMHL: A Multi-Modal Dataset for Detecting Human- and LLM-Generated Misinformation in the Medical Domaincited by 9
  36. 2023Automated Evaluation of Personalized Text Generation using Large Language Modelscited by 9
  37. 2023P5: Plug-and-Play Persona Prompting for Personalized Response Selectioncited by 8
  38. 2024MisinfoEval: Generative AI in the Era of "Alternative Facts"cited by 6

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Evaluation of Multilingual LLMs Personalized Text Generation Capabilities Targeting Groups and Social-Media Platformscited by 0
  2. 2026The Shibboleth Effect: Auditing the Cross-Lingual Distributional Skew of Large Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026To Lie or Not to Lie? Investigating The Biased Spread of Global Lies by LLMs
  2. 2025LLMs are Vulnerable to Malicious Prompts Disguised as Scientific Language
  3. 2025Whose Personae? Synthetic Persona Experiments in LLM Research and Pathways to Transparency
  4. 2026Persona Prompting as a Lens on LLM Social Reasoning
  5. 2024Learn and Unlearn: Addressing Misinformation in Multilingual LLMs
  6. 2024A Synthetic Dataset for Personal Attribute Inference
  7. 2025What Should LLMs Forget? Quantifying Personal Data in LLMs for Right-to-Be-Forgotten Requests
  8. 2024A Survey on the Honesty of Large Language Models