Mindgard is an offensive-security company for AI, offering automated red teaming and Dynamic Application Security Testing for AI (DAST-AI). Its platform continuously and automatically simulates attacks across the AI lifecycle to identify and help remediate vulnerabilities that only become detectable at runtime, such as prompt injection, jailbreaks, data extraction, and model evasion.
Founded in 2022 by Dr. Peter Garraghan and Steve Street, Mindgard spun out of Lancaster University in the UK, drawing on more than a decade of academic AI security research. The platform targets the unique attack surface of large language models and other AI systems, which traditional application security tools were not built to test.
In January 2025, Mindgard announced an $8M funding round led by .406 Ventures, with participation from Atlantic Bridge, WillowTree Investments, IQ Capital, and Lakestar. The capital funds R&D and expansion into the North American market, with a leadership presence established in Boston.
Mindgard serves security teams, AI/ML engineering teams, and red teams at enterprises deploying AI who need continuous assurance that their models are robust against adversarial attacks. It has won recognition including Best Cybersecurity Startup at the 2025 Cybersecurity Excellence Awards.
The company sits at the overlap of AI security and AI safety, addressing the growing need to test and harden production AI systems against an evolving threat landscape.