ServiceNow has expanded its AI Control Tower offering with a broader set of controls designed to govern AI systems, agents, models, workflows, and identities across enterprise estates, including environments outside ServiceNow’s own platform.
Announced at Knowledge 2026, the update pushes AI Control Tower beyond visibility and management into a wider operating layer spanning discovery, observability, governance, security, and financial measurement. ServiceNow said the product is intended to give customers a single place to manage AI regardless of where it runs.
ServiceNow said discovery now covers AI assets deployed across the organisation through 30 new integrations spanning AWS, Google Cloud, Microsoft Azure, and enterprise applications including SAP, Oracle, and Workday. It is also extending discovery to non-human identities and connected devices, bringing OT and IoT assets into the same governance model as cloud services and software agents.
On observability, the company is using capabilities from its recently completed Traceloop acquisition to monitor agent behaviour at runtime, including how agents reason, where they make decisions, and when they need intervention.
Governance and security are being widened at the same time. ServiceNow said AI Control Tower now includes five new risk frameworks aligned to NIST and the EU AI Act, extending policy controls beyond agents to cover models, prompts, datasets, and traditional machine-learning artefacts. On the security side, the integration of Veza technology brings identity graphing, scoped permissions, and least-privilege enforcement to AI systems and connected assets.
The company said the platform can detect when an agent exceeds its permissions and shut it down in real time. It is also adding cost tracking and ROI dashboards as part of a measurement layer designed to monitor AI spend and value as deployments expand. Jon Sigler, executive vice president and general manager of AI Platform at ServiceNow, said: “Enterprises are under real pressure to deploy AI and show results, but there’s a major gap between adoption and accountability.”
Availability is staggered. ServiceNow said features in the Australia release are rolling out from April 2026, with AI Agent Advisor and Intelligent Approvals generally available in May. The new AI Control Tower enhancements enter Innovation Lab in May, with general availability expected in August 2026. The rollout gives customers a period to test the controls in lower-risk settings before wider deployment.
As companies move from pilots and copilots to mixed estates of models, agents, cloud services, and workflow engines, the architecture around AI is changing. Model choice remains important, but the operational layer around identity, runtime visibility, policy enforcement, permissions, cost control, and auditability is becoming part of the same technology buying decision. ServiceNow is using AI Control Tower to place itself at that layer.
The company’s recent moves support that direction. On the same day as the AI Control Tower announcement, ServiceNow said it had passed $1 billion in AWS Marketplace transactions and outlined a deeper architecture linking AI Control Tower with Amazon Bedrock AgentCore, alongside new workflow integrations for security, IT operations, and telecommunications. Over recent months it has also added identity and asset intelligence through its Veza and Armis acquisitions. In separate announcements this week, ServiceNow said Autonomous Security & Risk had been launched to govern every AI agent, identity, and connected asset, while an update on the Armis acquisition described machine identities as outnumbering human identities by more than 80 to one.
Those additions point to a market in which AI governance, identity security, workflow automation, and infrastructure visibility are converging. ServiceNow already sits inside service management, operations, and enterprise workflow processes, and AI Control Tower is being built on the CMDB and Context Engine that map assets, services, people, and processes together. That gives the product a role inside day-to-day operational systems rather than on the edge of them.
There is also commercial backing for the platform strategy. ServiceNow reported first-quarter subscription revenue of $3.671 billion, up 22% year on year, and said the number of Now Assist customers spending more than $1 million in annual contract value had grown by more than 130% from a year earlier. The company also said its platform now carries more than 100 billion workflows annually. AI estates are growing more complex, and the centre of gravity is shifting towards the systems that can observe, govern, authorise, and measure what those agents do once they are deployed.
You must be logged in to post a comment.