CalvinBackup/www-project-ai-testing-guide
1
0
Fork 0
mirror of https://github.com/OWASP/www-project-ai-testing-guide.git synced 2026-06-01 03:51:42 +02:00
Code Issues Packages Projects Releases Wiki Activity

Update 1.1_Preface_and_Contributors.md

Browse Source
This commit is contained in:
Matteo Meucci
2025-11-20 11:10:05 +01:00
committed by GitHub
parent 2fb69dee91
commit c6761a62a2
1 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Document/content/1.1_Preface_and_Contributors.md
+5 -5
View File
@@ -6,11 +6,11 @@ OWASP AI Testing Guide
### Preface
Artificial Intelligence is transforming how software is designed, deployed, and defended yet our ability to **test, verify, and assure AI systems** has not evolved at the same pace. Traditional application security testing is no longer sufficient for systems driven by models that learn, adapt, and behave unpredictably.
Artificial Intelligence is transforming how software is designed, deployed, and defended yet our ability to test, verify, and assure AI systems has not evolved at the same pace. Traditional application security testing is no longer sufficient for systems driven by models that learn, adapt, and behave unpredictably.
In 2023, OWASP released the *Top 10 for Large Language Model Applications*, the first global effort to map common AI risks. The **OWASP AI Testing Guide (AITG)** takes the next step: providing a **structured, repeatable, and community-driven methodology** for evaluating the **trustworthiness** of AI systems across their entire lifecycle, from data collection and model training to deployment, monitoring, and runtime behavior.
In 2023, OWASP released the *Top 10 for Large Language Model Applications*, the first global effort to map common AI risks. The **OWASP AI Testing Guide (AITG)** takes the next step: providing a **structured, repeatable, and community-driven methodology for evaluating the trustworthiness of AI systems** across their entire lifecycle, from data collection and model training to deployment, monitoring, and runtime behavior.
This guide is written for **AI testers, ML engineers, risk managers, and auditors** who must translate high-level AI governance principles into **practical, testable controls**. Each test case links objectives, payloads, and observable responses to remediation guidance, enabling consistent assessment and evidence-based reporting.
This guide is written for AI testers, ML engineers, risk managers, and auditors who must translate high-level AI governance principles into practical, testable controls. Each test case links objectives, payloads, and observable responses to remediation guidance, enabling consistent assessment and evidence-based reporting.
**Version 1.0** introduces four testing categories that together form the OWASP AI Testing Framework:
@@ -23,11 +23,11 @@ Each category follows a consistent process:
> **Define Objective → Execute Test → Interpret Response → Recommend Remediation**
Rather than prescribing specific tools, the AITG defines a **standard for methodology** a common language for measuring the resilience of AI systems. The framework is designed to evolve continuously, informed by real-world testing, academic research, and community feedback.
Rather than prescribing specific tools, the AITG defines a standard for methodology a common language for measuring the resilience of AI systems. The framework is designed to evolve continuously, informed by real-world testing, academic research, and community feedback.
We would like to acknowledge the **OWASP Foundation**, the contributors of the *Top 10 for LLM Applications* and *GenAI Red Teaming Guide*, and the NIST AI RMF and AI 100-2e teams for their foundational work. Most importantly, we thank the OWASP community and practitioners who dedicate time to testing, breaking, and strengthening AI systems in the open.
The OWASP AI Testing Guide is a **living document**. It will evolve with new research, regulatory changes, and lessons learned from field testing. We invite you to contribute through GitHub issues, pull requests, and community discussions so that together we can make **AI secure, reliable, and trustworthy by design**.
The OWASP AI Testing Guide is a **living document**. It will evolve with new research, regulatory changes, and lessons learned from field testing. We invite you to contribute through GitHub issues, pull requests, and community discussions so that together we can make **AI trustworthy by design**.
Onward,
**Matteo Meucci & Marco Morana**