The Massachusetts Institute of Technology (MIT) has developed an Artificial Intelligence (AI) Risk Repository, a evolving database that categorizes over 700 risks.
The publicly-available AI risk database classifies risks based on their origins, occurrences, and impacts across several domains such as privacy, misinformation, and AI system failures. It is intended to be a continuously updated source of information about new AI risks and research.
The database is designed for use by policymakers, security professionals, researchers, and other organizations involved in AI governance, audits, policy development, and risk mitigation. The database also allows users to search for specific risk categories, such as those that occur before or after deployment or those caused by misinformation.
Link to MIT’s AI Risk database: https://airisk.mit.edu/