Come Join Us!
AVID is a community-driven effort, involving volunteer engineers, hackers, and researchers like you! The easiest way to be involved in AVID is to join our discord server. We follow the AI Village Code of Conduct in this server.
Also, here are a few specific ways you can support our work.
Submit a Report
You can use our Vulnerability Reporting Form to report an AI failure mode in the wild. The AVID team will review and integrate this report—with proper attribution to you—into a new or existing vulnerability in AVID.
You can also open a pull request to directly add a report or vulnerability to our GitHub.
We welcome comments, feedback, and inputs for our taxonomy and database structure. Please e-mail us with your comments and we’ll get back to you!
Our ongoing and planned research areas include but are not limited to
- Audits of open-source AI models and datasets
- Evaluation tooling specific to an area (e.g. fairness, privacy, security) or domain (e.g. NLP, CV, multimodal models)
- AI vulnerability research
If you’re interested in collaborating with us, please say hi in our discord, send us an e-mail, or fill out the Interest Form below.