Lasso Security

Lasso Security offers an AI security platform for enterprises building and deploying AI agents and LLM applications. According to its website, the platform connects discovery, risk management, automated red teaming, and runtime protection in one continuous loop so agentic applications stay within their intended scope.

What It Is

Lasso is an enterprise AI security product aimed at security teams that need visibility and control over agents, models, and AI applications. Its stated workflow is Discover, Assess, Protect, covering the lifecycle from inventory through runtime enforcement.

How the Platform Works

• Discovery & AI-BOM: discovers agents through platform integrations and homegrown applications through CI connections, mapping models, system prompts, tools and guardrails.
• AI Security Posture Management: analyzes misconfigurations, policy gaps, supply chain risk and exposure, with NIST and OWASP framework alignment.
• Automated AI Red Teaming: runs multi-turn agentic attacks from a library the site says exceeds 3,000 attack types, including context poisoning and tool chain manipulation.
• Runtime Enforcement: enforces policies inline at the proxy, API, or AI Gateway layer.
• AI Detection & Response: detects threats mapped to MITRE and OWASP and provides attack context and impact analysis.

Use Cases and Coverage

The site lists use cases including AI agent governance, agentic AI risk management, AI gateway security, AI coding assistants, Claude Enterprise, MCP security, and prompt injection protection, plus industry pages for public sector, healthcare, and finance. It also describes an Intent Security approach that analyzes the intent behind an agent's actions rather than fixed patterns.

Vendor Performance Claims

Lasso states 98.6% accuracy across content, context, and intent detections, under 50ms per classification, and that it is 570x more cost-effective than cloud-native guardrails. These are vendor-published figures. The company also announced CPU-based guardrails.