---
id: 1686
title: "We need a new way to measure AI security"
source_domain: blog.trailofbits.com
source_url: "https://blog.trailofbits.com/2023/03/14/ai-security-safety-audit-assurance-heidy-khlaaf-odd/"
published_at: "2023-03-14T12:00:47Z"
tags: ["ai-security", "security-audit", "machine-learning", "risk-assessment", "trail-of-bits", "assurance", "evaluation-benchmark"]
phying_url: "https://phying.com/item/1686"
lang: en
---

# We need a new way to measure AI security

## Summary

Trail of Bits launched an ML/AI assurance practice combining safety and security methodologies to build risk assessment frameworks and cybersecurity evaluation benchmarks for AI-based systems.

## Intelligence Metadata

- **Sources:** blog.trailofbits.com
- **Published:** 2023-03-14T12:00:47Z
- **Original:** https://blog.trailofbits.com/2023/03/14/ai-security-safety-audit-assurance-heidy-khlaaf-odd/
- **Phying News:** https://phying.com/item/1686