Guide for Contributors
Learn how to contribute to the AI Weakness Enumeration database and help secure AI systems worldwide.
What We're Trying to Do
The AI Weakness Enumeration (AWE) project aims to create a comprehensive, standardised database and classification system for AI vulnerabilities. Our goal is to help researchers, developers, policymakers, and organisations understand, track, and mitigate risks in AI systems.
By building a centralised repository with consistent taxonomy, we enable:
- Systematic analysis and comparison of AI security issues
- Better identification of patterns and trends in AI vulnerabilities
- Improved knowledge sharing across the AI security community
- More effective development of mitigation strategies
- Informed policy and decision-making around AI security
Your contributions are essential to building this resource. Every vulnerability you document helps make AI systems safer for everyone.
What Contributors Should Know
Submission Process
Submissions go through a review process to ensure quality and consistency:
- Fill out the submission form with as much detail as possible
- Our team reviews your submission for completeness and accuracy
- If needed, we may contact you for clarification or additional information
- Approved submissions are assigned an AWE ID and published in our database
What Makes a Good Submission
- Clear descriptions: Explain the vulnerability in terms that others can understand
- Relevant references: Include links to research papers, advisories, or documentation
- Impact assessment: Describe the potential consequences of the vulnerability
- Affected systems: Specify which models, frameworks, or systems are impacted
- Mitigation strategies: If known, include information about how to address the vulnerability
Getting Help
If you have questions about the submission process, need help filling out the form, or want to discuss a potential vulnerability before submitting, please don't hesitate to reach out. You can include questions in your submission notes, and our team will do their best to assist you.
Ready to Contribute?
Submit a vulnerability to help us build a comprehensive database of AI security issues.
Submit a Vulnerability