ServiceNow unveils Zurich platform to drive secure enterprise AI adoption

ServiceNow has launched its Zurich platform release, designed to accelerate enterprise artificial intelligence adoption with new secure, scalable, and governable features.

The latest release introduces several additions to ServiceNow's AI platform including Build Agent, Developer Sandbox, Autonomous Workflows with Agentic Playbooks, and new security consoles aimed at protecting data and governing system integrations at scale.

AI platform enhancements

ServiceNow's Zurich release aims to help organisations adopt AI across their operations with the introduction of Build Agent - a tool that supports rapid, production-ready app creation using natural language commands. Build Agent enables employees to describe their requirements conversationally, prompting the system to design, build, test, and integrate applications while maintaining security and auditability by default.

According to ServiceNow, this approach lowers the barriers for app development within enterprises and extends these capabilities to developers and so-called citizen creators, with every app featuring built-in compliance controls. The release also sees the introduction of Developer Sandbox environments, allowing teams to collaborate, test and iterate features without impact on live systems.

This platform update also adds capability for sandboxes to be versioned and used by multiple teams, enabling faster software delivery cycles and safer experimentation. According to ServiceNow, this structure helps reduce rework, shortens feedback loops, and supports rapid, lower-risk deployment of higher-quality applications.

Security and governance

To address concerns around security in AI deployments, the Zurich release introduces the ServiceNow Vault Console and Machine Identity Console. The Vault Console is designed to help organisations identify, classify, and protect sensitive data across workflows. It provides dashboards for compliance monitoring and makes recommendations for securing newly discovered sensitive information.

The Machine Identity Console focuses on managing access and authentication for bots and APIs. Platform teams can gain visibility into all inbound integrations, monitor for weak or outdated credentials, and take remediation steps directly from the console. This aims to help organisations maintain strong security as their volumes of programmatic access and AI agents increase.

Security features in the Zurich platform build upon the previously launched AI Control Tower, which offers centralised oversight and lifecycle management for AI agents, models and workflows, providing embedded compliance and governance for agentic AI systems.

"Zurich marks a turning point for enterprise AI. ServiceNow is delivering multi-agentic AI systems in production that are not just powerful, but governable, secure, and built for scale," said Amit Zavery, President, Chief Operating Officer, and Chief Product Officer at ServiceNow. "We are transforming the enterprise tech stack to be AI-native - from autonomous workflows that act on data with precision, to developer tools that democratise high-velocity innovation. With built-in controls for security, risk, and compliance, we're helping organisations move beyond experimentation and into a new era of intelligent execution."

Autonomous workflows

The Zurich platform release also incorporates agentic playbooks, supporting autonomous workflows where artificial intelligence and human oversight cooperate. Playbooks have typically involved automation based on pre-defined processes; agentic playbooks insert AI agents to automate decision-making and task execution, allowing human employees to intervene where needed.

ServiceNow cites an example within credit card support, where an agentic playbook could direct an AI agent to verify identity, freeze a card, and issue a replacement, while providing opportunities for human agents to participate if necessary. This approach is designed to maintain efficiency, consistency, and trust across workflows.

The unified platform also merges Process and Task Mining insights, aiming to provide organisations with end-to-end visibility over how work is completed. Insights gathered from these tools help to determine optimal points where AI agents can be deployed for maximum effect without undermining the need for human input where expertise is essential.

Gartner research is referenced by ServiceNow in the context of this release, stating, "By 2029, over 60% of enterprises will adopt AI agent development platforms to automate complex workflows previously requiring human coordination." Gartner also notes that agentic AI features will likely become near ubiquitous in enterprise software, transforming user experiences and business operations.

All of the features outlined in the Zurich platform release are available via the ServiceNow Store for organisations seeking to enhance their AI capabilities with built-in compliance and governance.