British cybersecurity startup Maze has raised $25 million in Series A funding to advance its AI-driven cloud security platform that deploys autonomous software agents to identify and remediate critical vulnerabilities in enterprise cloud environments.
The round was led by Theory Ventures, with participation from Cherry Ventures and Tapestry VC. The new funding brings Maze’s total capital raised to $31 million since its founding just nine months ago.
Unlike conventional vulnerability management tools that overwhelm security teams with exhaustive patch lists, Maze is building an AI-native security platform designed to automate the work of human analysts. Its system uses swarms of AI-powered agents that analyze cloud telemetry, mimic attacker behavior, and autonomously patch or report exploitable weaknesses.
The company said its agents break down workloads into thousands of concurrent tasks, enabling comprehensive, scalable testing of all possible attack paths in a customer’s cloud infrastructure. The approach is designed to focus on the small subset of vulnerabilities that are both exploitable and likely to lead to a breach.
“Maze can accurately identify the tiny number of vulnerabilities that are exploitable and likely to cause a breach, resolving them automatically,” the company said. “We’re building a comprehensive, AI-native security platform—starting with an agent for investigating, triaging, and resolving cloud vulnerabilities.”
The company reports that early deployments are underway across more than 10 organizations, including two Fortune 200 firms.
Maze plans to use the new funds to expand its engineering team and extend the agent framework to additional cloud security operations beyond vulnerability management. The startup’s bet—shared by its backers—is that the latest generation of large language models (LLMs) has matured to the point where autonomous software can reliably interpret and secure complex, idiosyncratic cloud environments.
The funding reflects growing investor confidence in the potential of AI agents to transform cybersecurity, particularly in automating high-volume, high-risk tasks that traditionally require human expertise.
