ServiceNow Inc. on Thursday introduced major advancements to its platform, commencing the third generation of its employee-centric artificial intelligence (AI) harness alongside a runtime control system named AI Passport.
The announcements are intended to democratize AI development across enterprise workforces while establishing strict, machine-speed safety boundaries for autonomous AI agents.
The updated platform centers on ServiceNow EmployeeWorks, an environment that enables non-technical workers to build, customize, and deploy AI agents without writing code. The agents are designed to execute routine daily tasks and navigate complex persona-based workflows autonomously across various channels and business applications.
To support technical teams, ServiceNow also introduced ToolSmith, an agentic developer environment embedded within EmployeeWorks. ToolSmith allows developers to quickly build and deploy specialized enterprise AI tools, a practice commonly referred to as vibe coding, reducing development timelines from days to minutes while maintaining enterprise-level security protocols.
“AI’s real promise is to close the gap between intention and action,” said Jacqui Canney, chief people and AI enablement officer at ServiceNow. “ServiceNow EmployeeWorks makes that shift possible by giving employees more time to solve customer problems that require human judgment.”
The platform’s underlying third-generation AI harness combines reasoning models with contextual employee data, governed workflows, and auditable actions. According to Bhavin Shah, senior vice president of ServiceNow’s employee experience group, the system operates as an enterprise operating system that connects workers to the full scope of ServiceNow’s platform on day one.
Alongside employee tools, ServiceNow unveiled AI Passport, a credentialing authority layer designed to enforce security and runtime control over autonomous AI agents and Model Context Protocol (MCP) servers.
Built on open specifications, AI Passport issues cryptographically signed credentials that agents refer to at runtime, including within external environments like NVIDIA Corp.’s OpenShell. Rather than relying solely on post-execution audit logs, the framework pre-authorizes what data, tools, and actions an agent can access, preventing unauthorized drift before it occurs.
If an agent exceeds its assigned boundaries, the system enforces graduated responses, ranging from context redirection to immediate machine-speed kill switches.
“As AI agents proliferate across the enterprise, visibility alone isn’t enough,” said Jon Sigler, executive vice president and general manager of AI Platform at ServiceNow. “ServiceNow closes that gap with an operational harness across AI Passport and new capabilities that unify agent authority, governance, and operational oversight so enterprises can scale agentic AI with control.”
The expanded operational suite also grants organizations centralized visibility to discover active agents, regulate multi-agent workflows, and track AI expenditure against concrete business outcomes. All execution data is stored within ServiceNow’s Super Intelligence Control Tower, maintaining an immutable audit trail for enterprise compliance.
