Geordie AI Raises $30M Series A as Enterprises Race to Secure AI Agents

Geordie AI has raised a $30 million Series A round led by Balderton Capital, as enterprises increasingly look for ways to govern and secure the rapidly expanding use of autonomous AI agents inside their organizations. The round also included participation from Crosspoint Capital, alongside existing investors General Catalyst and Ten Eleven Ventures. The latest raise brings the company’s total funding to $36.5 million.

The London-based startup is positioning itself at the center of a growing enterprise challenge: how to safely deploy AI agents that can independently make decisions, interact with systems, and execute tasks across sensitive corporate environments. While many organizations have embraced copilots and workflow automation tools, security teams are increasingly concerned about what happens when AI systems begin operating with greater autonomy and broader access permissions.

AI Agents Are Creating a New Security Layer Enterprises Must Manage

The rise of agentic AI has introduced a new category of cybersecurity and governance problems that traditional security tooling was not designed to address. Unlike conventional software, AI agents can dynamically chain actions together, access multiple systems, retain context, and modify their behavior based on changing objectives.

Geordie’s platform is designed to give enterprises visibility into those systems by mapping which agents exist, what data they can access, how they behave, and the risks they may introduce across corporate infrastructure. The company describes the platform as a “single source of truth” for AI agent governance.

The company’s growth suggests enterprises are taking the issue seriously. According to Geordie, annual recurring revenue grew 1,300% during the first five months of 2026 as organizations accelerated AI deployments.

Geordie has also quickly gained visibility within the cybersecurity industry, winning the 2026 RSAC Innovation Sandbox competition and appearing in Gartner’s Market Guide for Guardian Agents earlier this year.

Beam Focuses on Runtime AI Agent Remediation

A key component of Geordie’s platform is Beam, the company’s runtime remediation suite introduced earlier this year. Rather than relying solely on static rules or proxy-based controls, Beam uses what Geordie calls “context engineering” to continuously shape and constrain AI agent behavior while the systems are actively operating.

The system monitors how agents are configured, what actions they are attempting to perform, and the environments they interact with in real time. Based on that behavior, Beam feeds contextual controls and policy guidance back into the agent loop itself. The goal is to reduce risky or unintended actions without introducing the latency and operational bottlenecks commonly associated with gateway-based AI security architectures.

This approach reflects a broader shift happening across enterprise AI security. As organizations move beyond isolated pilots and begin deploying hundreds or even thousands of agents, security teams are increasingly searching for adaptive governance systems rather than traditional perimeter-based controls.

A Founding Team Built Around Cybersecurity and AI

Part of Geordie’s appeal to investors appears tied to the experience of its founding team. The company was founded in 2025 by former leaders from Darktrace and Snyk, two companies heavily associated with AI-driven cybersecurity and developer security tooling.

CEO Henry Comfort previously served as COO Americas at Darktrace, while co-founder Hanah-Marie Darley led security and AI strategy initiatives at the company. Co-founder Benji Weber previously held a senior engineering leadership role at Snyk focused on scalable security systems.

Balderton Capital partner James Wise, who will join Geordie’s board following the investment, framed AI agents as the next operational layer enterprises must secure. According to Wise, enterprises increasingly view agents as core infrastructure, but governance tooling has lagged behind adoption.

Pharmaceutical and Research Companies Are Early Adopters

One of Geordie’s early enterprise customers is Owkin, an AI-driven drug discovery and biomedical research company that operates hundreds of AI agents across more than 50 petabytes of data.

According to Geordie, an early proof-of-concept deployment at Owkin helped identify potential exposures that the customer internally estimated could have resulted in between $12 million and $13 million in risk.

The example highlights how industries dealing with highly sensitive intellectual property and regulated data may become some of the earliest large-scale adopters of AI agent governance platforms.

The Emerging Market for Agentic AI Security

The funding round underscores how quickly the AI security market is evolving beyond traditional application security and model protection. A growing number of startups are now focused specifically on securing AI agents, autonomous workflows, and tool-connected reasoning systems.

As enterprises expand the use of coding agents, research agents, workflow orchestrators, and autonomous copilots, concerns around permission sprawl, prompt injection, unintended actions, and invisible agent-to-agent interactions are becoming increasingly important for CISOs and compliance teams.

Geordie is betting that AI agent governance will become a foundational layer of enterprise infrastructure rather than a niche security category. The company plans to use the new funding to expand engineering teams, enhance its platform capabilities, and grow its US operations as demand for agentic AI security accelerates.