AI Models & Platforms
Ivanti Gives AI Clients Governed Access to Neurons Context

Key PointsIvanti's MCP server limits AI access to Neurons data through existing OAuth and…
View more
- Ivanti's MCP server limits AI access to Neurons data through existing OAuth and role-based permissions.
- Ivanti reports 99% active Copilot use across its organization within 90 days of internal deployment.
- Customers must weigh AI vendor contracts as part of governance since vendors control returned data handling.
Ivanti announced on November 18, 2026 a Model Context Protocol (MCP) connection that gives supported AI clients governed access to operational data in Ivanti Neurons, its system of record for endpoint, service and exposure information.
The connection is intended to ground AI recommendations, investigations and actions in device, digital-experience, patch, exposure and software intelligence. Ivanti’s MCP documentation lists GitHub Copilot, Anthropic Claude and Neurons agents as supported systems, while support for additional clients and custom agents can vary by release. MCP is an open standard created by Anthropic for connecting AI systems with data sources, tools, resources and prompts.
Existing Permissions Govern AI Access
The MCP server makes authorized Neurons data, tools, resources and prompts available according to configured permissions. OAuth, role-based access controls and policy enforcement continue to govern interactions, while existing platform controls maintain auditability and organizational oversight. Configured permissions therefore set the boundary of each AI-assisted workflow.
Neurons remains the system of record, and MCP functions as a protocol rather than a data store. Ivanti says it does not use customer data to train models.
Documented uses include finding failed patches and overdue scans, grouping devices by likely cause and identifying critical exposures for investigation. Other workflows surface abnormal endpoint signals, recent changes and fleet patterns, or summarize environment health, risk trends and the factors driving change.
Internal Copilot Adoption Informs Product Development
The product work follows internal deployments of Microsoft Copilot across the business and GitHub Copilot in engineering. Ivanti reports reaching 99% active Copilot use across the organization in 90 days, with 58% of employees now classified as Power Users and 64% of developers operating in the highest stage of its GitHub Copilot maturity model, where they use multi-agent workflows.
Ivanti also reports that engaged GitHub Copilot users merge 5.4 times more pull requests and merge them 1.9 times faster than their peers. The internal program combines employee training and governance with department-level think tanks and an AI center of excellence intended to turn experimentation into repeatable practices.
A companion announcement distinguishes native AI capabilities inside Ivanti products from MCP access for supported external systems. New assistive capabilities include device-query building, PowerShell stages, AI insights and recommendations, AI reporting and patch insights. Administrators retain control over whether and how recommended actions proceed.
Customer Zero Deployment and Reported Results
Ivanti validates Autonomous Endpoint Management in its own production environment under its Customer Zero approach, covering more than 10,000 internal endpoints for a distributed, predominantly remote workforce. It reports $11.2 million in total financial benefits and a 483% return on investment over three years, plus 66,000 manual tasks automated and 55,600 hours of employee productivity gained annually.
Those figures describe Ivanti’s own implementation. The Autonomous Endpoint Management product page identifies the financial-benefit, ROI and automation results as coming from a specific implementation, with customer results varying by environment, configuration and usage.
The MCP design creates a clear division of responsibility: Ivanti’s controls determine what an authorized AI system can retrieve, while the customer’s agreement with the AI vendor governs how returned data is handled. That makes the choice of AI client and its contractual terms part of the governance decision, alongside the permissions and data maintained in Neurons.












