paperJanuary 2025Unreviewed

OpenAI's Approach to External Red Teaming for AI Models and Systems

L. Ahmad, S. Agarwal, Michael Lampe, Pamela Mishkin

arXiv.org

Abstract

Red teaming has emerged as a critical practice in assessing the possible risks of AI models and systems. It aids in the discovery of novel risks, stress testing possible gaps in existing mitigations, enriching existing quantitative safety metrics, facilitating the creation of new safety measurements, and enhancing public trust and the legitimacy of AI risk assessments. This white paper describes OpenAI's work to date in external red teaming and draws some more general conclusions from this work.

Framework mappings

NIST AI Risk Management Framework

MEASUREMeasure

Suggested from the entry's categories.

Cite

@misc{ahmad2025openais,
  title = {{OpenAI's Approach to External Red Teaming for AI Models and Systems}},
  author = {L. Ahmad and S. Agarwal and Michael Lampe and Pamela Mishkin},
  year = {2025},
  month = jan,
  eprint = {2503.16431},
  archivePrefix = {arXiv},
  doi = {10.48550/arXiv.2503.16431},
  url = {https://www.semanticscholar.org/paper/4329ac5ac885b9bfe6510d98cfbde77806f6e82e}
}