What Is Elisity Network Security?

Key Takeaways:

IonQ agreed on October 12, 2026 to acquire Elisity in an all-stock deal valued at up to $1.3 billion, with up to $150 million tied to earnouts and closing expected in Q4 2026.

Elisity enforces identity-based microsegmentation through existing switches from Cisco, Arista, Juniper, HPE Aruba and Hirschmann, plus Palo Alto firewalls, without endpoint agents or new hardware.

An Omdia survey of 352 security leaders found 99% are implementing or planning microsegmentation, yet only 9% of those with active programs have covered 81% or more of critical systems.

Elisity’s August 2026 CLI exposes 466 commands and requires human approval on every state-changing action, a deliberate rejection of the standing MCP server model.

The St. Luke’s University Health Network deployment covered 85,000 medical devices across 15 hospitals in 46 days, according to the vendor’s announcement.

The Identity-First Approach to Segmentation

Elisity’s platform has three core components. IdentityGraph consolidates data from identity providers, endpoint tools, CMDBs, and IoT discovery into a single record per asset. The Dynamic Policy Engine then links policy group membership to that record, updating policies automatically when risk attributes change. The Virtual Edge operates outside the data path, programming policy directly into network hardware.

The IonQ Acquisition and the Quantum Angle

This enforcement method distinguishes Elisity from agent-based competitors. According to Elisity’s platform documentation, policy enforcement occurs through supported wired infrastructure from Cisco, Arista, Juniper, HPE Aruba, and Hirschmann, as well as Cisco Catalyst 9800 wireless controllers and Palo Alto Networks firewalls in Dynamic Address Groups. No endpoint agent is needed, and IP addresses remain unchanged.

Such a design is crucial for environments like hospitals and industrial sites, where devices often run locked firmware and cannot support agents. For these assets, the switch becomes the enforcement point. However, this approach depends heavily on hardware support. Devices outside the compatibility list leave the enforcement layer without a role.

The Open CLI and Customer-Built AI Agents

On August 5, 2026, Elisity unveiled an open command-line interface (CLI) allowing security teams to connect their own AI agents. The CLI includes 466 commands covering the Cloud Control Center API, with output formats like JSON, YAML, CSV, and tables. It supports profiles for production, staging, and lab environments. Every command that changes state pauses for human approval, and deletions require separate confirmation.

Elisity chose not to develop a Model Context Protocol (MCP) server. They argue that MCP introduces a privileged, persistent connection that could expand an agent’s access by default. This decision stems from concerns over vulnerabilities like CVE-2025-6514, disclosed in July 2025 and rated 9.6 out of 10, which allowed malicious MCP servers to execute OS commands on connecting machines. While the flaw’s severity is based on the vendor’s description, it highlights the risks of MCP’s default trust model.

# Example CLI workflow: discover, review, then approve a policy change.
# Commands are representative; consult support.elisity.com for exact syntax.

elisity login --profile prod
elisity devices list --unclassified --output json > unclassified.json
elisity policy simulate --group "Clinical-IoMT" --input proposed-policy.yaml
elisity policy apply --group "Clinical-IoMT" --input proposed-policy.yaml --require-approval

# Production deployment should include credential management, rate limiting, and change logging, not shown here.

Elisity’s built-in AI, called Elisity Intelligence, runs on private, single-tenant LLMs. It supports three functions: a conversational assistant, device classification, and policy recommendations. The company reports a 77% increase in customer usage of Elisity Intelligence between April and June 2026, mainly for policy work and device lookups. But these figures come from Elisity’s press release, with no independent verification. The human approval step limits automation benefits and could slow response times if bad recommendations occur.

The Deployment Gap Omdia Measured

A survey by Omdia of 352 security leaders at organizations with over 1,000 employees in healthcare and manufacturing reveals the market’s potential. Commissioned by Elisity, the study shows 99% of respondents are implementing or planning microsegmentation. Yet, only 9% of those with active programs have covered 81% or more of critical systems. Nearly half experienced lateral movement incidents in the past year. The gap between intent and execution highlights a significant opportunity for vendors like Elisity.

Zero Trust initiative	Implemented (Omdia, 352 respondents)
Multi-factor authentication	76%
Endpoint / extended detection and response	70%
Identity and access management	64%
Least-privilege access controls	36%
Microsegmentation of critical apps / systems	24%

The survey also asked which features matter most. Identity-based microsegmentation was top at 69%, followed by lateral movement prevention at 54% and quick deployment at 51%. Agentless deployment, which Elisity emphasizes, ranked near the bottom at 26%. Buyers want identity control but seem less willing to prioritize deployment ease, especially after some teams faced failed agent rollouts. This suggests that agentless methods are more attractive to those with prior bad experiences.

How This Compares to VLAN and Agent-Based Segmentation

The drawbacks of VLAN-based segmentation are well known. When St. Luke’s University Health Network in Pennsylvania and New Jersey tried traditional VLAN microsegmentation, they faced extensive IP address changes, vendor coordination, and manual VLAN management across hundreds of sites. The legacy approach simply couldn’t scale efficiently.

Agent-based solutions install enforcement software on each workload, which works for servers but fails on many medical devices and industrial controllers. Licensing costs are high, and policy authoring remains the main expense. Elisity shifts enforcement to switches and policy management to a correlation engine, reducing costs but narrowing reach.

h>
Requires IP changes

Approach	Enforcement point	Works on agentless devices
VLAN segmentation	Layer 2 / Layer 3 boundary	Yes	Yes
Agent-based microsegmentation	Endpoint kernel agent	No	No
Elisity identity-based	Existing network switches	Yes	No

Other vendors like Illumio, Guardicore (Akamai), and Cisco’s Tetration also compete in this space. But all face the same fundamental challenge: the enforcement layer itself can be an attack surface, as seen in vulnerabilities like Cisco’s recent security flaw. For more details, see our breakdown of network segmentation for security and cost realities for smaller teams.

The IonQ Acquisition and the Quantum Angle

The deal’s specifics include IonQ issuing shares based on a 20-day volume-weighted average price, between $33.07 and $49.61 per share, with up to $150 million in earnouts. Closing is expected in Q4 2026, pending regulatory approval.

IonQ sees enterprise challenges in containing fast-moving attacks and preparing for future threats from quantum decryption. Their shared roadmap aims to extend Elisity’s IdentityGraph with cryptographic inventory and risk data, linking cryptographic dependencies to identities and business processes. The company points to White House executive orders pushing federal agencies toward post-quantum cryptography as a sign of demand. But this roadmap remains a plan, not a product, and the companies will operate separately until deal closure.

IonQ reports that Elisity has over 100 enterprise deployments, doubled revenue and customer count year-over-year, and maintains zero churn among deployed customers. These claims are self-reported and unverified by independent audits. IonQ’s estimate of a $10 billion market for microsegmentation underscores their optimism.

What Customer Deployments Actually Show

The most concrete example is the St. Luke’s deployment. The health system secured more than 85,000 medical devices across 15 hospitals in just 46 days, with no outages or hardware changes, according to the vendor. Ransomware’s blast radius shrank from widespread to a single switch. They even enabled remote robotic surgeries previously blocked due to lateral movement risks.

Other clients include GSK, Main Line Health, MultiCare Health System, and Shaw Industries. MultiCare’s CISO Jason Elrod described how Elisity’s interface helps continuously validate policies against adversary techniques, surfacing policy drift within minutes rather than during incidents. These accounts are from Elisity’s own materials, without independent verification. The 46-day timeline is a vendor-reported figure for a single deployment, not a typical result.

For the regulatory context, see our HIPAA compliance analysis, which explains why healthcare networks face the tightest segmentation pressures.

What to Watch

The success of the IonQ deal hinges on three factors. First, whether it closes in Q4 2026 and the cryptographic roadmap delivers. Second, whether Elisity’s zero-churn claim holds after ownership transfer. Third, whether the CLI approach to AI remains viable as the industry shifts toward MCP despite its known security flaws. If open CLIs with human approval become industry standard, Elisity will have led the way. If MCP improves and CVE-2025-6514-like flaws are fixed, the CLI may be a temporary stopgap.

More broadly, the key question is whether identity-based segmentation via existing switches becomes the norm for the 90% of organizations lagging behind, as Omdia found. This approach avoids the two main failure modes of legacy projects (IP renumbering and agent deployment) but introduces hardware dependency. It’s better suited for healthcare and manufacturing than cloud-native environments, where agent and service-mesh methods still dominate.

More in-depth coverage from this blog on closely related topics:

Sources and References

Sources cited while researching and writing this article: