An AI blue team playbook

Mike Tan; Kureha Yamaguchi; Anna Raney; Victoria Nockles; Margaux Leblanc; Shiri Bendelac

doi:10.1117/12.3021908

7 June 2024 An AI blue team playbook

Mike Tan, Kureha Yamaguchi, Anna Raney, Victoria Nockles, Margaux Leblanc, Shiri Bendelac

Proceedings Volume 13054, Assurance and Security for AI-enabled Systems; 130540R (2024) https://doi.org/10.1117/12.3021908
Event: SPIE Defense + Commercial Sensing, 2024, National Harbor, Maryland, United States

Conference Poster

Abstract

In a fiercely competitive landscape, we are deploying AI systems faster than they can be security tested and defended. With developers under pressure to deliver on functionality and performance as quickly as possible, security is too often left as an afterthought. In response to emerging security challenges, we present a playbook to establish an AI blue teaming process for mitigating vulnerabilities before they can be exploited in the wild. By describing the process as part of a larger framework known as Build-Attack-Defend (BAD), we define an iterative and collaborative process between the AI system development and security teams, as well as various stakeholders. Our playbook contains the blue teaming historical context, process, lessons learned and hypothetical examples, serving as a starting point for embedding security at the heart of AI-enabled systems.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Mike Tan, Kureha Yamaguchi, Anna Raney, Victoria Nockles, Margaux Leblanc, and Shiri Bendelac "An AI blue team playbook", Proc. SPIE 13054, Assurance and Security for AI-enabled Systems, 130540R (7 June 2024); https://doi.org/10.1117/12.3021908

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available