TROJAI

AI is reorganizing our world like the web did in the early 90's. Like the web, AI is developed with technology that was not built with security in mind and we have been dealing with the consequences ever since. TrojAI protects AL/ML systems throughout the development pipeline from training through to deployment. Our tools can audit training data for embedded Trojan attacks, assess model robustness to long-tailed real-world edge cases, and filter inputs into deployed models with our AI Firewall. Our mission is to help global changemakers build a better world by advancing Trustworthy AI.