Mindgard

Mindgard is a cybersecurity company specialising in securing artificial intelligence systems. Founded in 2022 by Dr Peter Garraghan and Steve Street, the company spun out of Lancaster University, leveraging more than a decade of AI security research. Mindgard addresses security risks specific to AI, including LLM prompt injection and jailbreaks, which traditional security tools may not manage effectively.

Mindgard's core offering is an AI security platform that provides dynamic application security testing and automated red teaming across the AI lifecycle. The solution is model-agnostic and supports a range of AI systems, including generative AI, large language models, natural language processing, image and audio models. It integrates with CI/CD and MLOps pipelines, enabling security teams and developers to test for threats such as data leakage, model extraction and evasion during development and runtime. The company maintains a continuously updated library of AI attack scenarios developed through research-led security expertise.

Mindgard operates as a SaaS business serving organisations in sectors including financial services, healthcare, manufacturing and cybersecurity. Its platform is designed to help teams discover, assess and defend AI systems, agents and applications through automated security testing, vulnerability assessment and runtime protection.

Keywords: AI security, automated red teaming, DAST-AI, LLM security, generative AI security, adversarial machine learning, AI vulnerability management, continuous security testing, model security, prompt injection, AI risk management, cybersecurity, MLOps security, AI penetration testing