The malicious use of artificial intelligence: Forecasting, prevention, and mitigation
Apollo (University of Cambridge)2018
Citations Over Time
Miles Brundage, Shahar Avin, Jack Clark, Helen Toner, Peter Eckersley, Ben Garfinkel, Allan Dafoe, Paul Scharre, Thomas Zeitzoff, Bobby Filar, Hyrum S. Anderson, Heather M. Roff, Gregory C. Allen, Jacob Steinhardt, Carrick Flynn, Seán Ó hÉigeartaigh, SJ Beard, Haydn Belfield, Sebastian Farquhar, Clare Lyle, Rebecca Crootof, Owain Evans, Michael Le Page, Joanna J. Bryson, Roman V. Yampolskiy, Dario Amodei
Abstract
This report surveys the landscape of potential security threats from malicious uses of AI, and proposes ways to better forecast, prevent, and mitigate these threats. After analyzing the ways in which AI may influence the threat landscape in the digital, physical, and political domains, we make four high-level recommendations for AI researchers and other stakeholders. We also suggest several promising areas for further research that could expand the portfolio of defenses, or make attacks less effective or harder to execute. Finally, we discuss, but do not conclusively resolve, the long-term equilibrium of attackers and defenders.
Related Papers
- → GAN（Generative Adversarial Nets）(2017)21,735 cited
- → Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks(2017)21,465 cited
- → The global landscape of AI ethics guidelines(2019)4,566 cited
- → Superintelligence: paths, dangers, strategies(2015)2,004 cited
- → Agnostic Learning with Unknown Utilities(2016)1,383 cited