Identifying and Mitigating the Security Risks of Generative AI
Abstract
Every major technical invention resurfaces the dual-use dilemma -- the new technology has the potential to be used for good as well as for harm. Generative AI (GenAI) techniques, such as large language models (LLMs) and diffusion models, have shown remarkable capabilities (e.g., in-context learning, code-completion, and text-to-image generation and editing). However, GenAI can be used just as well by attackers to generate new attacks and increase the velocity and efficacy of existing attacks. This paper reports the findings of a workshop held at Google (co-organized by Stanford University and the University of Wisconsin-Madison) on the dual-use dilemma posed by GenAI. This paper is not meant to be comprehensive, but is rather an attempt to synthesize some of the interesting findings from the workshop. We discuss short-term and long-term goals for the community on this topic. We hope this paper provides both a launching point for a discussion on this important topic as well as interesting problems that the research community can work to address.
Cite
@article{arxiv.2308.14840,
title = {Identifying and Mitigating the Security Risks of Generative AI},
author = {Clark Barrett and Brad Boyd and Elie Burzstein and Nicholas Carlini and Brad Chen and Jihye Choi and Amrita Roy Chowdhury and Mihai Christodorescu and Anupam Datta and Soheil Feizi and Kathleen Fisher and Tatsunori Hashimoto and Dan Hendrycks and Somesh Jha and Daniel Kang and Florian Kerschbaum and Eric Mitchell and John Mitchell and Zulfikar Ramzan and Khawaja Shams and Dawn Song and Ankur Taly and Diyi Yang},
journal= {arXiv preprint arXiv:2308.14840},
year = {2024}
}