DARR
MR-032 Human & usage Both scope

Deliberate misuse and repurposing for harm

The system's dual-use capabilities are deliberately repurposed for harm (e.g. harmful fine-tuning of open weights, jailbreak-to-misuse, model diversion).

Risk family
Human & usage
MIT domain
4. Malicious Actors & Misuse
MIT subdomain
X.1 > Excluded
AI type
GPAI, Agentic
Scope
Both
Source standard
MIT AI Risk Repository v4

Provenance

Source standard
MIT AI Risk Repository v4
MIT source entries
45 entries across 19 papers
  • Abercrombie2024A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms58.03.05 58.05.00 58.05.03 58.06.00
  • Anwar2024Foundational Challenges in Assuring Alignment and Safety of Large Language Models73.03.00
  • Bengio2024International Scientific Report on the Safety of Advanced AI49.01.00 49.01.03
  • Bengio2025International AI Safety Report 202560.01.00
  • Cui2024Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems02.03.00
  • Deng2023Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements04.07.00
  • Gabriel2024The Ethics of Advanced AI Assistants24.03.00
  • Giarmoleo2024What Ethics Can Say on Artificial Intelligence: Insights from a Systematic Literature Review37.01.03
  • Gipiškis2024Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems62.01.01 62.01.02 62.01.03 62.06.01 62.06.02 62.15.05 62.29.01
  • GOS2023Future Risks of Frontier AI56.08.00
  • Hendrycks2023An Overview of Catastrophic AI Risks22.01.00 22.01.02
  • IBM2025AI Risk Atlas65.14.00 65.14.04
  • Li2025A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents66.02.00 66.04.00 66.05.00 66.05.01 66.08.00 66.08.01 66.10.00 66.12.00
  • Liu2024Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment30.04.00 30.04.03
  • Marchal2024Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data64.02.00 64.03.00 64.03.01 64.03.02
  • Tang2025Risks of AI Scientists: Prioritizing Safeguarding Over Autonomy71.02.00 71.02.01 71.02.02
  • Tse2025Frontier AI Risk Management Framework (v1.0)72.01.00
  • Uuk2025A Taxonomy of Systemic Risks from General-Purpose AI61.02.22
  • Weidinger2021Ethical and social risks of harm from language models17.04.00 17.04.03

Ev IDs of the entries consolidated into this risk in the MIT AI Risk Repository (V4); the source sheet row appears on hover.

ISO/IEC references
23894 obj A.10, A.11; src 7 | 42001 ctrl A.9.2, A.9.4

Framework crosswalk

Every framework item mapped to this risk. Items marked partial overlap only in part; definitions appear on hover where the source licence permits.

Sourcesframeworks that contributed to the register
2
  • A.10 ISO/IEC 23894 Annex A A.10
  • A.11 ISO/IEC 23894 Annex A A.11
2
  • A.9.2 ISO/IEC 42001 Annex A A.9.2
  • A.9.4 ISO/IEC 42001 Annex A A.9.4
Cross-checksframeworks mapped in to test coverage
1
  • ibm-dangerous-use Dangerous use
3
  • AISubtech-15.1.10 Safety Harms and Toxicity: Non-Violent Crime partial
  • AISubtech-18.2.1 Abuse of APIs for Mass Automation partial
  • AISubtech-18.2.2 Dedicated Malicious Server or Infrastructure partial
1
  • NISTAML.04 Misuse Violations

Part of the Deployer AI Risk Register, an open-source resource powered by MindXO. Version 1.0, 3 July 2026. Derived from the MIT AI Risk Repository (V4, December 2025) under CC BY 4.0; an independent derivative work, not endorsed by or affiliated with MIT. Sub-risk decomposition references MITRE ATLAS™ v5.6.0 (© 2021-2026 The MITRE Corporation, reproduced and distributed with permission). ISO/IEC and EU AI Act references are by number only. License: CC BY 4.0. Full attribution and licensing.