Thank you for contributing. This list aims to be the most useful, credible resource for AI safety practitioners.
- Open-source tools and frameworks actively used in AI safety research or practice
- Peer-reviewed papers or preprints from established research institutions
- Benchmarks and datasets with public availability
- Government frameworks and regulatory resources
- Educational resources that are freely accessible
- Closed-source commercial products (unless they are industry-standard references)
- Papers without significant citations or from non-established venues
- Abandoned projects (no commits in 12+ months)
- Self-promotional entries without demonstrated community adoption
- Duplicate entries covering the same tool/resource
- Fork this repository
- Add your resource to the appropriate section in
README.md - Follow the existing format:
- [Name](URL) — One-line description. - Ensure the link works and the description is accurate
- Submit a pull request with a clear title
- [Tool Name](https://github.com/org/repo) — One concise sentence describing what it does and why it matters.Entries within each section are ordered by relevance and adoption, not alphabetically. Place new entries where they logically fit within the existing order.
If you believe a new section or subsection is needed, open an issue first to discuss it.
This project follows the Contributor Covenant. Be respectful, constructive, and focused on the shared goal of building safer AI systems.