Cisco to Buy Astrix Security for Roughly $400 Million to Govern AI Agents and Non-Human Identities Across the Enterprise
The deal, announced May 4, brings the Israeli non-human-identity startup into Cisco's security platform alongside Identity Intelligence, Duo, Secure Access, and Splunk.
Editor's Note ·
- Clarification:
- Two of the article's six cited sources are not on the project's source_allowlist (blogs.cisco.com, astrix.security). Both are primary vendor sources for this acquisition: Cisco's own announcement and Astrix's co-founder post. Their claims were verified verbatim against the local snapshots during review. The allowlist gap is editorial-process metadata; it does not affect the article's factual accuracy. Adding both domains to config/source_allowlist.txt is recommended.
- Clarification:
- The article attributes the quote 'authenticate and authorize non-human identities under a strict Zero Trust model' to Cybersecurity News. That source returned HTTP 403 to the fetcher, so the quote cannot be verified from the local snapshot. The substance is corroborated by other allowlisted sources but the specific quotation cannot be confirmed.
- Clarification:
- The Bailey quote in the article reads 'Enable organizations to adopt AI securely and at scale' (capital 'Enable'). The verbatim source-1 (SiliconANGLE) text reads 'enabling organizations to adopt AI securely and at scale'. Same substance, different verb form.
Overview
Cisco announced on May 4, 2026 its intent to acquire Israeli cybersecurity startup Astrix Security, in a deal Calcalist reports is valued at roughly $400 million. Cisco itself did not disclose financial terms in its announcement, framing the acquisition as a way to extend its security platform to cover AI agents and other non-human identities that increasingly act on behalf of human employees.
In its official post, Cisco said the goal is “Securing the Agentic Workforce” by folding Astrix’s discovery, governance, and threat-detection tooling into Cisco Identity Intelligence, Cisco Secure Access, Duo, and the company’s Splunk SIEM. Astrix’s co-founders confirmed the same integration scope in a blog post on the company’s site.
What Astrix Does
Astrix was founded in 2021 by Alon Jackson and Idan Gour, both veterans of Israel’s elite Unit 8200. Jackson serves as Co-Founder and CEO and Gour as Co-Founder and CTO.
The company’s platform focuses on what the security industry calls non-human identities — the API keys, service accounts, and OAuth tokens that machines use to talk to other machines, plus the AI agents that increasingly piggyback on those credentials. According to SecurityWeek, Astrix provides “visibility into non-human identities and the activity of AI-driven agents, along with lifecycle management and automated detection and remediation of over-privileged, unnecessary, or malicious access — including compromised credentials and rogue agent behavior.”
The Astrix platform creates real-time inventories of AI agents, MCP servers, and non-human identities operating in production environments. Cisco’s own description groups the capabilities into four buckets: “Discovery & Governance for AI Agents,” “Agentic Access & Lifecycle Management,” “Agentic Threat Detection & Response,” and “Secrets Management” for keys and tokens across vaults and cloud environments.
Funding History
Before the Cisco deal, Astrix had raised approximately $85 million in venture funding, according to Calcalist. Its Series B in late 2024 brought in $45 million, led by Menlo Ventures through its Anthology Fund — a vehicle Menlo runs in partnership with Anthropic. Workday Ventures, Bessemer Venture Partners, CRV, and F2 Venture Capital also participated in that round.
Why Cisco Is Buying
The acquisition reflects Cisco’s bet that the next wave of identity security has to cover machines and AI agents, not just human users. Peter Bailey, SVP/GM of Cisco’s Security Business Group, framed the rationale around two statistics from Cisco’s own AI Readiness Index: only 24% of organizations can control agent actions with proper guardrails and live monitoring, and just 31% feel fully capable of securing their agentic AI systems. Cybersecurity News attributed the 24% figure to the same Cisco AI Readiness Index.
In a post announcing the deal, Astrix CEO Alon Jackson wrote that “The gap between agentic capability and organizational readiness is widening. Closing it requires the kind of platform, reach, and expertise that this combination brings.” Cisco’s Bailey said the goal is to “Enable organizations to adopt AI securely and at scale.”
SecurityWeek frames the move as Cisco extending “zero trust principles to the emerging ‘agentic workforce,’ where AI agents and machine identities are rapidly expanding the enterprise attack surface.”
Integration Path
Astrix’s product is slated to plug into multiple existing Cisco offerings. The Astrix blog post lists the integration scope as Cisco Identity Intelligence, Secure Access, Duo, and Splunk — the same set Cisco itself names in its announcement. SecurityWeek’s account notes integration with Cisco’s identity intelligence, secure access, and Duo IAM without explicitly listing Splunk. Cybersecurity News describes the goal as letting customers “authenticate and authorize non-human identities under a strict Zero Trust model.”
What We Don’t Know
- The official deal price. Cisco did not disclose financial terms; the $400 million figure comes from Calcalist’s reporting and is corroborated by SecurityWeek. SiliconANGLE notes that The Information had earlier speculated a figure of around $300 million.
- Closing date. None of the surveyed sources gave a target close.
- Astrix’s headcount or the post-close team structure inside Cisco’s Security Business Group.
- Whether Astrix’s standalone product will continue to be sold to customers who do not run Cisco’s platform.
Why It Matters
Non-human identity security has become one of the more contested segments of enterprise cybersecurity as agentic AI moves from demos into production. Buying Astrix gives Cisco an off-the-shelf inventory and policy layer for AI agents and machine credentials, and ties that layer into the identity, access, and SIEM products customers already run. For the broader market, it is the largest publicly reported acquisition focused specifically on AI agent governance to date — and a signal that the platform vendors intend to absorb this layer rather than let it remain a standalone category.