Securing AI: How Traditional Vulnerability Disclosure Must Adapt

Abstract

Like traditional software, vulnerabilities in machine learning software can lead to sabotage or information leakages. Also like traditional software, sharing information about vulnerabilities helps defenders protect their systems and helps attackers exploit them. This brief examines some of the key differences between vulnerabilities in traditional and machine learning systems and how those differences can affect the vulnerability disclosure and remediation processes.

Reducing the Risks of Artificial Intelligence for Military Decision Advantage

“…28 Third, fixing ML vulnerabilities often creates other problems. 29 To address vulnerabilities in ML systems, developers have to retrain the system so that it is no longer susceptible to that deception. Retraining is not only expensive, but also has diminishing returns in terms of addressing the underlying issue.…”