AI Red Teaming for Image Models
With this update, Mindgard’s platform and CLI have been updated to support image models.
The award received by the deep-tech startup Mindgard underscores the critical importance of AI security and highlights the innovative research capabilities of UK universities.
London (6 June 2024) – AI Red Teaming provider Mindgard has been honored as the "UK's Most Innovative Cyber SME 2024" at Infosecurity Europe 2024, a prestigious European event for the cybersecurity community. Mindgard was selected as the winner among 14 finalists, recognizing its groundbreaking contributions to identifying and mitigating vulnerabilities in AI, GenAI, and LLMs.
As the adoption of AI and GenAI accelerates, many businesses overlook the significant security that could disrupt their operation, even state-of-the-art AI models are susceptible to attacks such as data leakage, prompt injection, and bypassing content harm systems. Traditional security tools often fail to detect these threats due to their complexity and the specialized skill set required to identify them. Current penetration testing methods are often costly, time-consuming, and quickly become outdated with model updates, leaving organizations exposed to potential risks within their AI models.
Mindgard, a deep-tech startup based in London, addresses these challenges. Founded in 2022 by Professor Peter Garraghan, alongside scientists and engineers from Lancaster University's cybersecurity lab, Mindgard has quickly established itself as a leader in AI security. Specializing in cybersecurity for AI, Mindgard empowers enterprises to secure their AI, GenAI and LLMs by swiftly identifying and mitigating vulnerabilities. Prof. Garraghan foresaw the potential for AI systems to transform business and society as early as 2016 and dedicated six years to researching and solving AI cyber risks.
Additionally, Mindgard has recently launched the Mindgard AI Security Labs, a free tool designed to empower enterprise security teams to deploy AI and GenAI securely. This is the world's most advanced Red Teaming platform, allowing engineers to conduct comprehensive penetration testing on their models. It guides them through simulations of various potential attacks, including poisoning, evasion, extraction and jailbreak. By enabling continuous security testing across a broad spectre of threats, it minimizes AI cyber risks, accelerates AI adoption, and unlocks the commercial value of AI/GenAI for businesses.
Infosecurity Europe is renowned for its commitment to identifying the most creative and original cybersecurity companies in the UK. For the ninth consecutive year, the competition sought to spotlight the industry's leading innovators, culminating in the selection of Mindgard as this year’s winner.
Mindgard joins a cohort of leading UK cyber SMEs including ANGOKA (2023), TryHackMe (2022), CAPSLOCK (2021), Hack the Box (2019), KETS Quantum Security (2018), Check Recipient (now trading as Tessian) (2017), Device Authority (2016) and Geolang (2015).
Peter Garraghan, Co-founder and CEO of Mindgard, commented: "This award is a testament to the outstanding work of our team and the research capabilities of UK universities in creating innovative technology. There is a growing concern within the tech community about AI security and a need for effective solutions. Mindgard’s mission is to translate cutting-edge research into an enterprise-grade platform that empowers organizations to test, detect, and respond to AI attacks effectively."
About Mindgard
Mindgard is a cybersecurity company specializing in security for AI.
Founded in 2022 at world-renowned Lancaster University and is now based in London, Mindgard empowers enterprise security teams to deploy AI and GenAI securely. Mindgard’s core product – born from ten years of rigorous R&D in AI security – offers an automated platform for continuous security testing and red teaming of AI.
In 2023, Mindgard secured $4 million in funding, backed by leading investors such as IQ Capital and Lakestar.
Thank you for taking the time to explore our article.
