French cybersecurity startup White Circle has successfully raised $11 million in seed funding from a group of angel investors, including executives from prominent tech firms like OpenAI and Google. Established in 2025, White Circle has created a pioneering AI control platform that serves as a protective layer for AI applications, addressing risks such as hallucinations and data leakage while enforcing organizational policies. This innovative platform results from the founder’s experience in exposing vulnerabilities in existing AI safeguards, highlighting its necessity for real-time monitoring and risk detection in enterprise settings. The new funding will facilitate rapid product development and expansion of White Circle’s customer base across the US, UK, and Europe.
Denis Shilov: Denis Shilov is the founder and CEO of White Circle, an engineer based in Paris who previously developed a viral universal jailbreak prompt inspired by a crime thriller that exposed safety vulnerabilities in major AI models. He is leading the development of runtime safety and alignment infrastructure tailored for enterprise AI deployments. Recent coverage highlights his vision for a purpose-built platform to monitor and shape AI behavior without compromising security or compliance.
White Circle: White Circle is a Paris-based cybersecurity startup offering a control layer for AI that unifies safety, security, evaluations, and performance optimization by monitoring inputs and outputs for risks such as hallucinations, prompt injections, harmful content, abuse, and model drift. The platform enforces organization-defined policies to secure AI agents, prevent data leakage, block malicious actions, and improve model accuracy over time with multilingual support. It recently announced seed funding from AI industry leaders to accelerate product development, hire talent in the US, UK, and Europe, and grow its customer base.
Founder Origin: Stemmed from founder’s viral jailbreak technique that demonstrated flaws in safeguards of leading AI models, prompting the need for production-grade monitoring.
Seed Investors: Backed by executives from OpenAI, Anthropic, DeepMind, Hugging Face, and Google including Durk Kingma, Guillaume Lample, Thomas Wolf, François Chollet, and Paige Bailey.
Market Positioning: Fills gap as the first platform purpose-built for real-time AI behavior monitoring, risk detection, and policy enforcement in enterprise settings.