An independent organization that thoroughly tests AI systems for flaws, bugs, and vulnerabilities, offering an easy to understand 'bug score' for any AI system out in use. Additionally, the company provides 'off switch' protocol information available to the public to decentralize safety.