British startup Maze, a platform that makes use of AI brokers to analyze and resolve cloud safety vulnerabilities, in the present day introduced a €21.8 million Sequence A spherical to proceed rising its workforce and broaden into new use instances for its AI-native safety platform.
The spherical was led by Concept Ventures, with participation from present traders Cherry Ventures and Tapestry VC. Coming lower than a yr after the corporate’s founding and 9 months after a €5.2 million Seed spherical led by Cherry Ventures, this brings complete funding to €27 million.
“Earlier than beginning Maze, we talked to a whole bunch of safety professionals and heard the identical criticism time and again – coping with vulnerabilities is a nightmare,” stated Harry Wetherald, Maze CEO and Co-founder. “Safety groups are already struggling in the present day, and as attackers get their fingers on AI, issues are solely going to get harder. Whereas AI provides us the flexibility to construct higher safety instruments, it is usually enabling unhealthy actors to ship complicated assaults at scale. Our purpose is to offer safety groups the instruments they should battle again.”
Based in 2024, Maze presents itself as a brand new sort of safety platform that makes use of AI brokers to cease cloud safety breaches earlier than they occur, beginning with an agent for investigating, triaging, and resolving vulnerabilities.
Maze’s founding workforce contains Harry Wetherald, Adrian Jozwik, and Santiago Castineira, who beforehand led product, design, and engineering groups at Tessian, Elastic, and Amazon. Along with a veteran workforce of ex Meta, Adyen, and Nvidia engineers, they constructed Maze’s first AI agent by replicating the workflow of professional human analysts.
The brokers have been deployed in manufacturing with over ten organisations, together with two of the Fortune 200, and quite a few high-growth know-how corporations.
“Maze’s product is nearly too good to be true. It’s a really novel use of Agentic AI that goes effectively past automating guide workflows,” stated Mandy Andress, CISO at Elastic.
Vulnerabilities are a quickly rising ache level for contemporary safety groups. In 2024, vulnerability exploitation increased by 34%, overtaking phishing amongst high causes of safety incidents. The variety of recognized vulnerabilities (CVEs) can also be on the rise, rising by nearly 40% in 2024. Given that the majority enterprises can only fix about 10% of their open vulnerabilities every month, and with AI now making attackers simpler, safety groups want new options.
Maze’s AI brokers are designed to function and suppose like people, as a substitute of counting on pre-defined logic and guidelines. By deploying hundreds of brokers to analyze their buyer’s cloud knowledge, Maze can precisely establish the tiny variety of vulnerabilities which are exploitable of their setting and prone to trigger a breach, resolving them routinely.
Beginning with an agent for investigating, triaging, and resolving cloud vulnerabilities, Maze is growing a complete, AI-native safety platform.
“To maintain tempo with attackers, safety instruments will have to be re-written to be AI-native,” stated Andy Triedman, Associate at Concept Ventures and newest addition to the Maze Board. “Whereas in the present day’s vulnerability administration instruments are constructed to combination knowledge for human analysts, Maze is designed from the bottom up for autonomous investigations and triage.”
