A Safe Harbor for AI Evaluation and Red Teaming
Abstract
Independent evaluation and red teaming are critical for identifying the risks posed by generative AI systems. However, the terms of service and enforcement strategies used by prominent AI companies to deter model misuse have disincentives on good faith safety evaluations. This causes some researchers to fear that conducting such research or releasing their findings will result in account suspensions or legal reprisal. Although some companies offer researcher access programs, they are an inadequate substitute for independent research access, as they have limited community representation, receive inadequate funding, and lack independence from corporate incentives. We propose that major AI developers commit to providing a legal and technical safe harbor, indemnifying public interest safety research and protecting it from the threat of account suspensions or legal reprisal. These proposals emerged from our collective experience conducting safety, privacy, and trustworthiness research on generative AI systems, where norms and incentives could be better aligned with public interests, without exacerbating model misuse. We believe these commitments are a necessary step towards more inclusive and unimpeded community efforts to tackle the risks of generative AI.
Cite
@article{arxiv.2403.04893,
title = {A Safe Harbor for AI Evaluation and Red Teaming},
author = {Shayne Longpre and Sayash Kapoor and Kevin Klyman and Ashwin Ramaswami and Rishi Bommasani and Borhane Blili-Hamelin and Yangsibo Huang and Aviya Skowron and Zheng-Xin Yong and Suhas Kotha and Yi Zeng and Weiyan Shi and Xianjun Yang and Reid Southen and Alexander Robey and Patrick Chao and Diyi Yang and Ruoxi Jia and Daniel Kang and Sandy Pentland and Arvind Narayanan and Percy Liang and Peter Henderson},
journal= {arXiv preprint arXiv:2403.04893},
year = {2024}
}