The Pulse
Vanta signs agreement to acquire Pillar for AI agent controls
Vanta has signed a definitive agreement to acquire AI security company Pillar Security. The companies plan to combine Pillar’s runtime protections and testing with Vanta’s trust-management platform, with features scheduled to arrive from Ja
AI.info Team · Written with AI · fact-checked ·
Vanta has signed a definitive agreement to acquire Pillar Security, bringing the AI security company’s runtime platform into Vanta’s governance offering. The agreement is for a planned acquisition; Vanta describes the combined product as a system for taking action on AI governance while agents are running.
In its November 9 announcement, Vanta said the deal is intended to connect runtime protection and adversarial testing to its Trust Graph. Vanta describes the Trust Graph as a live map of a customer’s trust program, powered by more than 400 integrations.
Connecting security tools and governance
Vanta says the products address a gap between AI security tools that detect threats and GRC systems that hold business context in policies and controls. The combined system will use that context to turn governance requirements into boundaries agents must follow.
The platform will block an agent attempting to access restricted data or share sensitive financial information in an unapproved channel. Vanta’s Agent will then alert owners and produce an audit trail of the protections applied.
Vanta’s announcement describes agents as moving beyond answering questions to tasks such as pushing code, writing customer records and acting across business systems. The announcement argues that governance based on periodic reviews may not match the pace of agents’ actions, making controls that operate at runtime part of the planned offering.
Features scheduled through summer 2027
The announcement lays out a staged product schedule, beginning with AI Inventory and runtime monitoring and detection in January 2027. AI Inventory is planned to map applications, agents, models and MCP servers, linking them to owners, approved uses, data access, vendors, risks and controls.
Runtime monitoring and detection, also scheduled for January 2027, is intended to identify suspicious activity and guardrail violations, with investigations prioritized using Trust Graph context. Red teaming is scheduled for spring 2027, with testing tailored to a customer’s business context to find vulnerabilities and exploitable paths.
Runtime enforcement is scheduled for summer 2027 and is intended to block unauthorized actions before they happen. AI Governance by Vanta is planned as a paid offering in January 2027.
Pillar’s research work
Pillar also brings an AI security research team that tracks emerging threats and develops guardrails and red-teaming techniques, Vanta said. The team developed SAIL, a framework for securing AI agents that Vanta says is aligned with ISO and NIST standards and OWASP guidance.
More than 20,000 customers use Vanta’s platform to earn and prove trust, and customers are asking for real-time AI control, Vanta says. IDC Research Director Phil Harris said governance needs controls that act in real time as agents operate, rather than systems that document risk only after the fact.