OWASP Machine Learning Security Top 10 (2023 edition)
The OWASP Machine Learning Security Top 10 project aims to provide an overview of the top 10 security risks associated with machine learning systems. This draft release (v0.3) is intended for developers, machine learning engineers, security practitioners, and application security experts.
Key features:
- Identifies and ranks the most critical security vulnerabilities in ML systems.
- Covers a range of threats, including adversarial attacks and non-adversarial scenarios.
- Focuses on security hygiene in ML operational and engineering workflows.
Target audience:
- Developers building ML-powered applications.
- ML engineers responsible for deploying and maintaining ML models.
- Application security experts securing ML systems.
- Operational practitioners involved in ML workflows.
Use cases:
- Security risk assessment of ML systems.
- Guidance for secure development and deployment of ML models.
- Awareness and training for ML security best practices.
