Pillar offers a unified AI security platform that automatically discovers and inventories all AI assets—models, agents, servers, and developer tools—across an enterprise, mapping their permissions, data sources, and integrations to produce a continuous risk posture score. It validates defenses with multi‑turn red‑team simulations and enforces adaptive runtime guardrails that block unauthorized tool calls, data leaks, and policy violations, while providing governance dashboards, audit logs, and compliance reporting for security and compliance teams.
Funding
$9M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
1OFounders
Product
Problem
Enterprises deploying custom AI agents, LLM‑powered workflows, and AI‑enabled tools lack visibility into what agents exist, what data and permissions they have, and how they interact, creating hidden attack surfaces such as shadow AI, prompt injection, data exfiltration, and privilege escalation.
Solution
Pillar provides a unified AI security platform that discovers and inventories all AI assets—including models, agents, MCP servers, and endpoint tools—across an organization. It maps each asset’s permissions, data sources, and tool integrations to generate a continuous risk posture score. The platform then validates defenses through multi‑turn red‑team simulations that expose real‑world attack paths, and it enforces adaptive runtime guardrails that block unauthorized tool calls, data leaks, and policy violations. Integrated governance features produce audit logs and compliance reports, enabling security and compliance teams to monitor, remediate, and demonstrate AI security posture at scale.
Target Audience
Primary customers are security and compliance teams in large enterprises that develop or operate internal AI agents, LLM‑driven applications, and AI‑enabled developer tools, as well as DevOps groups responsible for AI model deployment and governance.
Features
- Automated discovery of AI agents, models, MCP servers, and endpoint coding assistants, including hidden “shadow AI” assets
- Permission and data‑access mapping that identifies excessive scopes, hard‑coded secrets, and unauthorized PII handling
- Continuous risk scoring and change detection for new agents, permission changes, and tool connections
- Red‑team engine that executes multi‑turn adversarial attacks across agentic workflows to expose lateral movement and exfiltration paths
- Adaptive runtime guardrails that enforce policy per agent, block prompt injection, tool poisoning, and unauthorized data egress in real time
- Taint analysis tracing sensitive data from source to destination and automatically blocking illicit transfers
- Governance dashboard with audit logs, compliance reporting (GDPR, CCPA) and policy lifecycle management
- Integration hooks for AI gateways and CI/CD pipelines to embed security checks without redesigning existing stacks