迈向可信赖的AI开发：支持可验证声明的机制
计算机与社会
2020-04-22 v2
摘要
随着近期人工智能（AI）进展的浪潮，人们日益意识到AI系统的大规模影响，并认识到产业与学术界现有法规与规范不足以确保负责任的AI开发。为了让AI开发者赢得系统用户、客户、公民社会、政府及其他利益相关者对其负责任构建AI的信任，他们需要做出可验证的声明并为此负责。组织外部人员也需要有效手段来审查此类声明。本报告提出了不同利益相关者可采取的各类步骤，以改进关于AI系统及其相关开发过程所发表声明的可验证性，重点在于提供有关AI系统安全性、安全性、公平性与隐私保护的证据。我们为此分析了十种机制——涵盖机构、软件与硬件——并提出了旨在实施、探索或改进这些机制的推荐建议。
引用
@article{arxiv.2004.07213,
title = {Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims},
author = {Miles Brundage and Shahar Avin and Jasmine Wang and Haydn Belfield and Gretchen Krueger and Gillian Hadfield and Heidy Khlaaf and Jingying Yang and Helen Toner and Ruth Fong and Tegan Maharaj and Pang Wei Koh and Sara Hooker and Jade Leung and Andrew Trask and Emma Bluemke and Jonathan Lebensold and Cullen O'Keefe and Mark Koren and Théo Ryffel and JB Rubinovitz and Tamay Besiroglu and Federica Carugati and Jack Clark and Peter Eckersley and Sarah de Haas and Maritza Johnson and Ben Laurie and Alex Ingerman and Igor Krawczuk and Amanda Askell and Rosario Cammarota and Andrew Lohn and David Krueger and Charlotte Stix and Peter Henderson and Logan Graham and Carina Prunkl and Bianca Martin and Elizabeth Seger and Noa Zilberman and Seán Ó hÉigeartaigh and Frens Kroeger and Girish Sastry and Rebecca Kagan and Adrian Weller and Brian Tse and Elizabeth Barnes and Allan Dafoe and Paul Scharre and Ariel Herbert-Voss and Martijn Rasser and Shagun Sodhani and Carrick Flynn and Thomas Krendl Gilbert and Lisa Dyer and Saif Khan and Yoshua Bengio and Markus Anderljung},
journal= {arXiv preprint arXiv:2004.07213},
year = {2020}
}