HiddenLayer
Enterprise AI security platform — discovery, supply chain, runtime defense, and attack simulation.
What it does
The most comprehensive AI security platform on the market. Four integrated modules: AI Discovery (shadow AI inventory), AI Supply Chain Security (model scanning, AIBOM, Model Genealogy), AI Runtime Security (prompt injection, jailbreak, data leakage defense), and AI Attack Simulation (continuous automated red teaming). Gartner Cool Vendor. Backed by M12 (Microsoft), IBM Ventures, Booz Allen Ventures, Capital One Ventures.
Security relevance
Purpose-built for AI-specific threats — not retrofitted from traditional cybersecurity. Non-invasive architecture means no access to raw data or model weights required. Covers both GenAI and agentic systems including MCP-based workflows. Runtime module detects and blocks prompt injection, model extraction, and unauthorized tool use in real time. Supply chain module generates AIBOMs and tracks model lineage. OWASP GenAI Security Project sponsor.
When to use it
Deploy when you need a unified AI security platform that spans the full lifecycle from model discovery through production monitoring. Enterprise-grade with native integrations for AWS Bedrock, SageMaker, Databricks Unity Catalog, and major SIEM/SOAR tools. Significant procurement and integration commitment, but eliminates the need for multiple point solutions.
OWASP coverage
Risks addressed — mapped to both OWASP Top 10 standards. 5 in LLM, 5 in Agentic.
The raw record
What Yuntona stores. Single source of truth — fork it on GitHub.
name: HiddenLayer slug: hiddenlayer type: Mixed category: AI Guardrails & Firewalls url: https://hiddenlayer.com reviewed: 2026-04 added: 2026-04 updated: 2026-04 risks: llm: [LLM01, LLM02, LLM03, LLM04, LLM06] asi: [ASI01, ASI02, ASI04, ASI05, ASI10] complexity: Enterprise Only pricing: — audience: Blue Team lifecycle: [deploy] tags: [Enterprise, Platform, Runtime, Supply Chain]