TrojAI was founded in 2020 to protect AI and machine learning systems from adversarial attacks. The company provides a security platform designed to secure AI models and applications throughout the AI lifecycle, addressing both build-time and runtime security requirements.

The TrojAI platform operates in two primary modes. At build time, it automatically red teams AI models to identify vulnerabilities and safeguard model behavior before deployment. At runtime, the platform functions as a firewall that protects enterprises from threats and active attacks in real time, monitoring AI applications as they operate in production environments.

The company targets enterprise organizations deploying AI-enabled applications, particularly those operating in complex environments where AI security risks are expanding. TrojAI addresses the growing attack surface created by AI/ML deployments, focusing on mitigating risks and threats to both AI models and applications.

The platform is designed to integrate security into every stage of the AI lifecycle, enabling organizations to deploy AI systems with confidence while maintaining security controls. TrojAI offers services through technology partnerships, consulting partnerships, and channel partnerships to help organizations implement AI security measures. The company also provides AI Red Team Report Card assessments to evaluate AI security posture.