FireTail AI Security Posture Management (ASPM) is a platform that provides organizations with visibility into their AI usage, enabling discovery, risk assessment, and governance of AI models, agents, and integrations across the enterprise. The platform begins with automated discovery and inventory of all AI activity, including embedded LLMs and shadow AI instances, building a complete map of an organization's AI landscape. This inventory is then assessed for risks across three tiers: - **Code analysis:** Static analysis of AI and LLM integrations to identify security issues in coding and API-driven implementations. - **Log analysis:** Normalization and analysis of logs from multiple LLM providers to detect PII leakage, anomalies, and guardrail invocations. - **External LLM scanning (optional):** Intentional submission of malicious prompts to test AI implementations for susceptibility to abuse. Risk assessment covers prompt injection vulnerabilities, data leakage via logs or outputs, over-permissive API access, agent behavior deviations, and lack of guardrails or policy enforcement. Risks are ranked by business impact and linked to business context such as department, application, or regulatory exposure. The platform includes a governance engine that allows organizations to define AI usage policies, enforce role-based access, and apply output filtering guardrails. It provides a single-pane-of-glass dashboard for monitoring AI data flows and risk posture. FireTail ASPM also addresses compliance requirements, helping organizations manage regulatory risks related to GDPR, CCPA, and PII data handling in AI workflows.