HiddenLayer, an AI security startup, has raised $100 million in a Series B round led by Delta-v Capital, with participation from Ten Eleven Ventures, Morgan Stanley, M12, Booz Allen Hamilton and others, signaling strong investor interest as enterprises increasingly secure AI deployments.

The funding round aims to expand HiddenLayer’s AI security platform and scale its sales, distribution, and geographic reach.

Management stressed that, while integrations with larger platforms may come later, the immediate focus is scaling sales and expanding into Europe and the broader EMEA region.

Recent milestones include the August 2026 launch of Agent Harness Security as a standalone solution, the March 2026 expansion of agentic capabilities within Runtime Security, and August 2026 selection to support the US Department of Energy’s Prometheus AI security initiative for nuclear energy.

HiddenLayer highlights a strong research arm with 39 granted patents and 65 pending patents on adversarial detection, model protection, and AI threat analysis, and cites the 2026 AI Threat Landscape Report showing 96% of organizations view AI as critical, with about one-third unsure whether they’ve experienced an AI-related breach.

Industry context notes incumbent cybersecurity players may favor acquisitions over in-house development, while startups like Noma and Zenity compete in overlapping spaces.

HiddenLayer claims over a 10x increase in annual recurring revenue in the past year, with more than 50 new platform customers across finance, government, technology, and defense, plus international growth and involvement with a frontier model provider reaching 700 million weekly users.

The product scope has broadened to cover prompt injection, agent manipulation, and malicious tool use, extending beyond Discovery and Runtime Protection to address GenAI and agentic AI workflows.

Leadership envisions AI infrastructure governance features—discovery, identity, and policy controls—becoming more important as HiddenLayer scales with AI and expands into broader cybersecurity areas reliant on AI.

A key focus is protecting against open-source model risks, with HiddenLayer scanning around 50 AI file frameworks to verify model integrity and prevent hidden or misrepresented models from being used.

Funding will accelerate three focus areas: Agentic Runtime Security to monitor and halt AI agent misuse, Agent Harness Security to secure coding agents at runtime, and a new product to secure AI coding agents.

HiddenLayer plans to deepen channel relationships, accelerate international expansion, and extend its enterprise platform including regulatory and governance capabilities.