Black-Box Access is Insufficient for Rigorous AI Audits-Reference-Cited by-同舟云学术

Black-Box Access is Insufficient for Rigorous AI Audits

Published:2024-06-03 Issue: Volume:35 Page:2254-2272
ISSN:
Container-title:The 2024 ACM Conference on Fairness, Accountability, and Transparency
language:
Short-container-title:

Author:

Casper Stephen¹^ORCID,Ezell Carson²^ORCID,Siegmann Charlotte³^ORCID,Kolt Noam⁴^ORCID,Curtis Taylor Lynn¹^ORCID,Bucknall Benjamin⁵^ORCID,Haupt Andreas¹^ORCID,Wei Kevin²^ORCID,Scheurer Jérémy⁶^ORCID,Hobbhahn Marius⁶^ORCID,Sharkey Lee⁶^ORCID,Krishna Satyapriya⁷^ORCID,Von Hagen Marvin¹^ORCID,Alberti Silas⁸^ORCID,Chan Alan⁹^ORCID,Sun Qinyi³^ORCID,Gerovitch Michael³^ORCID,Bau David¹⁰^ORCID,Tegmark Max³^ORCID,Krueger David¹¹^ORCID,Hadfield-Menell Dylan¹^ORCID

Affiliation:

1. Massachusetts Institute of Technology, United States of America

2. Harvard University, USA

3. Massachusetts Institute of Technology, USA

4. University of Toronto, Canada

5. Centre for the Governance of AI, United Kingdom

6. Apollo Research, United Kingdom

7. Harvard University, United States of America

8. Stanford University, USA

9. Centre for the Governance of AI, Mila (Quebec AI Institute), Canada

10. Northeastern University, USA

11. University of Cambridge, United Kingdom

Publisher

ACM

Link

https://dl.acm.org/doi/pdf/10.1145/3630106.3659037

Reference350 articles.

1. The Grey Hoodie Project: Big Tobacco, Big Tech, and the Threat on Academic Integrity

2. Abubakar Abid, Mert Yuksekgonul, and James Zou. 2022. Meaningfully debugging model mistakes using conceptual counterfactual explanations. In International Conference on Machine Learning. PMLR, 66–88.

3. Julius Adebayo, Michael Muelly, Ilaria Liccardi, and Been Kim. 2020. Debugging tests for model explanations. arXiv preprint arXiv:2011.05429 (2020).

4. Openxai: Towards a transparent evaluation of model explanations;Agarwal Chirag;Advances in Neural Information Processing Systems,2022

5. AI Safety Summit. 2023. The Bletchley Declaration by Countries Attending the AI Safety Summit. https://www.gov.uk/government/publications/ai-safety-summit-2023-the-bletchley-declaration/the-bletchley-declaration-by-countries-attending-the-ai-safety-summit-1-2-november-2023

Cited by 2 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Mapping the landscape of ethical considerations in explainable AI research;Ethics and Information Technology;2024-06-25

2. Building a Culture of Safety for Ai: Comparisons and Challenges;2024