London-based AI security startup Overmind has secured €2.3 million (£2 million) in Seed funding to build what it describes as the supervision layer for agentic AI systems. The company will use the capital to expand its technical team, accelerate product development and scale go-to-market efforts across regulated industries including legal, healthcare and FinTech.
The round was led by specialist cybersecurity investor Osney Capital, with participation from 14Peaks, Portfolio Ventures, Antler and Endurance Ventures.
Founded in 2025, Overmind is positioning itself at the intersection of AI infrastructure, cybersecurity and governance, targeting one of the fastest emerging risks in enterprise AI adoption: how to safely deploy autonomous AI agents in live production environments.
Securing behaviour, not just models
While much of the AI security industry has focused on model robustness and adversarial inputs, Overmind argues that the real vulnerability lies elsewhere.
“The AI security industry is trying to secure the wrong thing,” says Tyler Edwards, co-founder and CEO of Overmind. “Models will always be vulnerable to adversarial inputs. But what happens when an agent is live in production, interacting with real systems, and its behaviour starts to drift? Right now, most teams have no idea.”
Overmind’s platform operates at the deployment layer, monitoring how AI agents behave once connected to enterprise systems. Using what it calls pattern-of-life analysis, the system observes agent interactions in real time, identifies behavioural deviations and intervenes before damage occurs.
The company says its infrastructure provides complete visibility into agent activity, detecting anomalies, preventing unsafe actions and maintaining compliance with regulatory and data privacy requirements.
Intelligence-grade leadership
Overmind’s founding team brings together experience from British intelligence and high-growth technology companies.
CEO Tyler Edwards spent eight years building AI systems for UK intelligence agencies including MI5, MI6 and GCHQ. His background in national security AI systems informs the company’s focus on mission-critical deployment, behavioural monitoring and adversarial resilience.
CTO Akhat Rakishev previously led machine learning infrastructure at Monzo and Lyst, overseeing scalable AI systems in high-traffic production environments. Chief Revenue Officer Sam Brunt has led go-to-market scaling at three unicorn companies: Funding Circle, Pipe and Vertice.
The combination of intelligence-grade AI expertise, production ML infrastructure and enterprise scaling experience is central to Overmind’s strategy of becoming a foundational security layer for agentic AI.
From supervision to continuous improvement
Beyond monitoring and intervention, Overmind integrates reinforcement learning to improve agent performance over time. By analysing production data, the platform feeds insights back into optimisation loops, allowing agents not only to remain secure but to become more accurate and specialised.
The company describes itself as a developer tool that enables AI systems to learn safely from real-world behaviour. This dual focus on security and performance aims to remove one of the key barriers to enterprise AI adoption.
According to Gartner estimates cited by the company, up to 40 percent of agentic AI projects could be cancelled by 2027 due to inadequate risk controls. Overmind argues that without a robust supervision layer, organisations will hesitate to deploy autonomous systems at scale.
Investor backing amid rising AI security momentum
Overmind’s €2.3 million Seed round arrives amid growing European investment in agentic AI infrastructure and governance technologies. Across 2025 and 2026, more than €26 million has flowed into startups addressing AI agent safety, API security and AI-driven security operations.
Adam Cragg, Partner at Osney Capital, says: “In the new frontier of autonomous AI, agent security, performance and execution are the ultimate competitive advantages. Overmind provides businesses with differentiated technology that monitors and secures agentic AI while iteratively improving model performance.”
Adam French, Partner at Antler, adds that Overmind is addressing one of the most critical bottlenecks in AI advancement: the safe supervision of autonomous agents in production.
Targeting regulated industries
With its Seed capital, Overmind plans to scale adoption in sectors where compliance, auditability and data privacy are paramount. Legal services, healthcare providers and FinTech firms are early focus areas, as these industries stand to benefit significantly from agentic AI while facing strict regulatory obligations.
As agentic AI systems increasingly move from experimentation into operational environments, the threat landscape is shifting. Traditional security tools were not designed for autonomous systems capable of executing tasks, accessing sensitive systems and making decisions without human oversight.
Overmind aims to define the security standard for this new era, offering infrastructure that observes, secures and continuously improves AI agents once they are live in production.
With intelligence-sector roots and venture backing from cybersecurity-focused investors, the company is positioning itself as a core trust layer for the safe deployment of autonomous AI across Europe and beyond.
Also Read: Deep Space Energy Raises €930K to Power Europe’s Lunar Ambitions with Nuclear Batteries