The deal targets non-human identity and AI agent security, as Cisco says only 24% of organisations have guardrails and live monitoring for agent actions.
Cisco plans to fold Astrix's tools for discovery, lifecycle management, threat detection and secrets management into Identity Intelligence, Secure Access and Duo products.
The company said the acquisition supports its wider push into AI security and zero-trust controls as enterprises deploy more autonomous agents with access to data and systems.
As AI agents begin creating other agents, how can we prevent an unstoppable cascade of digital identity theft?
Is Cisco building a unified fortress against AI threats or just a complex maze for its customers?
When a 'secure' AI agent inevitably causes catastrophic harm, who is ultimately held responsible?
Cisco’s $350M Acquisition of Astrix Security: Securing 100% of Non-Human Identities in the AI Era
Overview
In early 2026, Cisco acquired Tel Aviv-based Astrix Security for $250-350 million to address growing security challenges from AI adoption and interconnected systems. Astrix’s innovative agentless technology secures Non-Human Identities (NHIs) like AI agents and automated processes, filling a critical gap in Cisco’s AI security portfolio. This acquisition follows Cisco’s earlier moves, including the 2024 Splunk acquisition, and sets the stage for integrating Astrix’s capabilities with Cisco’s Duo, SecureX, and Galileo platforms. The integration aims to automate threat detection, accelerate safe AI adoption, and enhance compliance, positioning Cisco strongly amid fierce competition and increasing regulatory scrutiny in the rapidly consolidating AI security market.