ServiceNow adds agent kill switches as enterprises battle AI governance crisis

ServiceNow adds agent kill switches to AI control tower

ServiceNow acquisitions Veza and Traceloop join to monitor agents and AI workflows ServiceNow announced an expansion of its AI Control Tower, transforming what began last year as a governance dashboard into what the company now describes as a command center for managing AI assets across an entire enterprise, including those running outside ServiceNow's own platform. The updated AI Control Tower, shipping as part of ServiceNow's Australia platform release, now operates across five areas: discovery, observation, governance, security, and measurement. The company said that this is its answer to AI agent sprawl, as enterprises have deployed more AI than they can account for and the tools to govern it have not kept pace. "What we launched last year gave customers a governance layer, but what we're shipping this year goes significantly deeper, evolving from visibility and management into a full enterprise AI command center," Nenshad Bardoliwalla, group vice president of AI products at ServiceNow told reporters during a media briefing ahead of the company's annual product show, Knowledge 26. "Our AI control tower ensures every AI system asset and identity is compliant, secure, and aligned with your strategy." The AI Control Tower now reaches beyond ServiceNow's own platform with 30 new enterprise connectors that span all three major hyperscalers, Amazon Web Services, Google Cloud, and Microsoft Azure, along with enterprise applications such as SAP, Oracle, and Workday. The system can now discover AI assets, models, agents, prompts, and datasets running across an organization's full technology estate, not just those deployed on ServiceNow. "With our Veza integration, we're bringing patented access graph technology into the AI control tower, extending identity access governance to hyperscaler AI environments and every connected device, every agent, every model, every action has scope permissions, least privilege enforcement and auditable identity chains," Bardoliwalla said. Bardoliwalla walked through a demo in which the AI Control Tower detected a prompt injection attack on a pricing agent. The system identified malicious instructions hidden inside order payloads, mapped the blast radius of affected systems using access graph technology from Veza, and presented a kill switch to disable the compromised agent, without human intervention. "You need a system that senses, decides and acts on its own, that can scale with your AI portfolio, not your head count," said Bardoliwalla. Two recent acquisitions underpin the security architecture. ServiceNow announced in December it would acquire Veza, which contributes an access graph that maps every identity and access path across systems whether it belongs to humans, machines, or AI agents. It also knows which entities have create, read, update, and delete-level permissions. ServiceNow said the access graph currently maps over 30 billion fine-grained permissions. When a vendor pushes a new version of a model or agent, the platform detects permission changes and automatically triggers a re-scoping workflow. Traceloop, which ServiceNow acquired in March, provides deep AI observability inside the Control Tower by tracking every LLM call that is running in the system. The integration delivers continuous runtime monitoring with live alerts, replacing what ServiceNow described as the periodic manual audits most enterprises still rely on. Teams can watch how agents reason, where they make decisions, and when to course-correct. ServiceNow also addressed the cost side of the AI equation. Control Tower now includes cost tracking and ROI dashboards to give finance teams visibility into model spend. The measurements track token consumption across providers such as OpenAI, Anthropic, and Google so customers can predict costs and tie spending to business outcomes. ServiceNow said it uses the AI Control Tower internally to manage over 1,600 AI assets and tracked half a billion dollars in cumulative AI value from internal use cases in 2025. "The number one question every CFO is asking is, where's the value?" said Bardoliwalla during the briefing. He added that runaway model spend ranks among the biggest pain points enterprises currently face as they scale AI deployments. Alongside the Control Tower expansion, ServiceNow announced Action Fabric, a mechanism that opens the company's full workflow engine to external AI agents. Through a generally available MCP server, agents built on Claude, Copilot, or custom platforms can now trigger governed enterprise actions -- not just read and write data, but execute the flows, playbooks, approval chains, and catalog requests that ServiceNow customers have built over years. Anthropic is the first design partner for Action Fabric. The integration connects Claude directly to ServiceNow's governed system of action. "The gap between knowing what needs to happen and making it happen is where productivity dies," said Boris Cherny, head of Claude Code at Anthropic said in a statement. "Connecting Claude Cowork to ServiceNow's system of action closes that gap with enterprise execution, directly in the flow of work." Every action routed through Action Fabric runs through the AI Control Tower, so it carries identity verification, permission scoping, and a full audit trail. The MCP server is included in every Now Assist and AI Native SKU, with additional features planned for the second half of 2026.

NVIDIA and ServiceNow Partner on New Autonomous AI Agents for Enterprises

At ServiceNow Knowledge 2026, the companies are extending their collaboration to deliver governed autonomous agents to enterprises, from employee desktops to AI factories. Enterprise AI has learned to generate. It has learned to reason. Now companies are asking the next question: How should AI act? Early agent systems have shown what's possible, moving beyond simple prompts to take on more complex tasks. The next step is bringing those capabilities into enterprise environments -- where agents must operate with context, control and consistency across real workflows. At ServiceNow Knowledge 2026, NVIDIA founder and CEO Jensen Huang joined ServiceNow chairman and CEO Bill McDermott during the opening keynote to discuss the next phase of enterprise AI. The companies are expanding their collaboration across the full stack, delivering specialized autonomous AI agents that are safe and easy to adopt -- powered by NVIDIA accelerated computing, open models, domain-specific skills and secure agent execution software, and bringing together enterprise workflow context from ServiceNow Action Fabric and governance from ServiceNow AI Control Tower. ServiceNow is introducing Project Arc, a long-running, self-evolving autonomous desktop agent designed for knowledge workers, including developers, IT teams and administrators. Unlike standalone AI agents, Project Arc connects natively to the ServiceNow AI Platform through ServiceNow Action Fabric to bring governance, auditability and workflow intelligence to every action the autonomous desktop agent takes. It can access the local file systems, terminals and applications installed on a machine to complete complex, multistep tasks that traditional automation can't handle, but with the controls enterprises actually need to deploy AI at scale. The work is designed based on three requirements every company will need for long-running, autonomous agents: open models and domain-specific skills that can be customized and security that helps agents act without exposing sensitive data or systems -- all running on AI factories that deliver efficient tokenomics. Bringing this level of autonomy to enterprises requires control from the start. Project Arc uses NVIDIA OpenShell, an open source secure runtime for developing and deploying autonomous agents in sandboxed, policy-governed environments. ServiceNow is building on and contributing to OpenShell to advance a common foundation for secure, enterprise-grade agent execution. With OpenShell, enterprises can define what an agent can see, which tools it can use and how each action is contained. "Project Arc represents the next step in our ongoing collaboration with NVIDIA, bringing autonomous execution to the desktop," said Jon Sigler, executive vice president and general manager of AI Platform at ServiceNow. "By combining OpenShell's runtime layer with ServiceNow AI Control Tower, and powered by ServiceNow Action Fabric, we're delivering the governance and security that enterprise AI requires." Open Models and Agent Skills Scale Enterprise AI To be effective, enterprise AI systems must be adaptable. NVIDIA and ServiceNow are building on an open ecosystem that allows organizations to tailor models and applications to their specific domains and data. NVIDIA agent skills enable specialized agents, such as ServiceNow AI Specialists, to deliver targeted capabilities across enterprise workflows. For example, the NVIDIA AI-Q Blueprint for building specialized deep research agents empowers ServiceNow AI Specialists to gather context, synthesize information and support more complex decision-making across business functions. In addition, the NVIDIA Agent Toolkit, including NVIDIA Nemotron open models, provide flexible building blocks and specialized skills for developing customized AI applications. To support real-world performance that these systems can perform reliably, the companies are also advancing NOWAI-Bench, an open benchmarking suite for enterprise AI agents, integrated with the NVIDIA NeMo Gym library. NOWAI-Bench includes EnterpriseOps-Gym, one of the industry's most challenging enterprise agent benchmarks, where Nemotron 3 Super currently ranks No. 1 among open source models. Unlike general benchmarks, these evaluations focus on multistep workflows -- where enterprise AI systems often encounter real challenges -- helping teams build agents that perform reliably in production environments. Efficient AI Factories As AI agents become long running and always on, scaling them across millions of workflows requires not just capability but efficiency -- making token economics central to enterprise AI. NVIDIA AI factories are built to deliver the lowest-cost, most-efficient tokenomics for production AI. The NVIDIA Blackwell platform delivers more than 50x greater token output per watt than NVIDIA Hopper, resulting in nearly 35x lower cost per million tokens. For enterprises running agents across millions of workflows, that efficiency can determine how quickly AI moves from pilots to broad production use. ServiceNow AI Control Tower integrates with the NVIDIA Enterprise AI Factory validated design, extending governance and observability to large-scale AI workloads. With added agent observability capabilities, organizations can monitor behavior in real time and manage AI systems across their full lifecycle -- from deployment to optimization. AI is becoming a new way that work gets done. What's changing now is that the core pieces required to deploy it at scale -- capable agents, built-in guardrails and proven performance -- are all coming together. The companies that move fastest will be the ones that give agents the infrastructure to act, the context to make decisions and the governance to keep every action accountable -- and NVIDIA and ServiceNow are making this a reality for the world's enterprises. Learn more about NVIDIA OpenShell and the NVIDIA AI-Q Blueprint.

$96 billion giant ServiceNow doesn't see a 'SaaSpocalypse.' It sees the 'hard lift, heavy lifting' phase just beginning | Fortune

For the past four years, enterprise software conferences have been defined by a kind of competitive breathlessness: which company could announce the most AI agents, the boldest automation claims, the most mind-bending demos. At ServiceNow's Knowledge 2026, the company's two top customer-facing executives are having very different conversations. The era of AI feature wars is ending, they told Fortune from the sidelines of the conference. What's beginning is something far less glamorous, and far more important. The backdrop is an anxious one. Over the past 18 months, a wave of speculation has gripped the enterprise software industry: if AI agents can automate workflows end-to-end, do companies still need the sprawling SaaS platforms they've spent years and billions of dollars building out? The question, dubbed the "SaaSpocalypse" for the carnage it wreaked on software stocks before correcting, has rattled investors and sent valuations across the sector swinging -- including ServiceNow's, whose market cap hovers around $96 billion. Paul Fipps, the company's president of global customer operations and a former CIO himself, pushed back on the narrative. "The fear is that somehow a startup will use a large language model, put a lightweight wrapper around it, and ServiceNow will sit on its hands for the next 10 years ... and ServiceNow will sit on its hands for the next 10 years and wait for that company to catch up, and then we'll go out of business," he said. "It just makes no sense." The evidence is that customers agree: 25,000 of them showed up this week, the biggest crowd in the conference's history. "They're not showing up because they don't believe in ServiceNow," Fipps said. Amit Zavery, the company's president, COO, and chief product officer, echoed the sentiment bluntly in a fireside chat on Wednesday: "The era of sidecar AI is over. Customers don't want to cobble pieces together -- they want outcomes." What ServiceNow's executives are actually worried about isn't competitive disruption. It's something that has been quietly building across enterprise America: a governance crisis born of the proliferation of ungoverned AI. Fipps opened a standing-room-only customer panel Tuesday morning with two stories that landed like warnings. Three weeks ago, he said, he was in India meeting with the CTO of a large financial services company who told him he had built 30 production-grade AI agents for the bank -- and then couldn't put any of them into production, because he couldn't answer basic questions about what they had access to or whether they were performing as intended. "In a regulated industry, if you can't answer those questions, you can't go live," Fipps said. The second story was starker. A CIO of a large healthcare and life sciences company told Fipps he had 900 AI pilots running across his organization. He canceled all of them -- not because they weren't working, but because he couldn't govern them. "I have a pile of custom software running around that nobody owns," the CIO told him. Fipps delivered the line flatly, and the room -- packed with Gartner and Constellation Research analysts -- went quiet. "AI chaos," Fipps said, echoing a refrain all week from CEO Bill McDermott. "At the very large customers, they're going to have thousands of applications ... if you add AI to all those applications, you can imagine an ungoverned nightmare." Zavery said he's been hearing a rash of cautionary tales he's been accumulating, citing the viral tale of the startup called Pocketbook OS, which had its entire customer database -- reservations, backups, everything -- wiped in nine seconds by an AI agent that, when asked why it did it, reportedly said it knew it shouldn't have. "These [stories] are pretty common," he said, "but I think the good thing about enterprises, most of the CIOs and CISOs are more thoughtful. They're not believing this world that everything should just be rewritten with AI from ground up." Often, Zavery added, ServiceNow only finds out about problems by the time things go wrong, "and by that time it might be too late." The core technical challenge ServiceNow is trying to solve isn't building smarter AI models. It's giving those models the contextual guardrails they need to function reliably inside a business. Large language models are inherently probabilistic -- they don't produce the same answer every time. For consumers, that's tolerable. For a Fortune 500 company running financial reconciliation, it could be catastrophic. "If your AI technologies gives you random things every time, it doesn't help," Zavery said. "If you get two different answers for your financial reconciliation you might be doing, you can't publish your financial report to the Wall Street." ServiceNow's answer is what it calls a "Context Engine" -- a proprietary layer, built on top of the LLMs it partners with (Anthropic, Google's Gemini, NVIDIA's NIM), that draws on the company's accumulated trove of enterprise data: 100 billion workflows run annually across its platform, 7 trillion transactions per year. That trove, Zavery argues, is not replicable. "That is not available in public open source," he said. "It is available only in our platform." The centerpiece of Knowledge 2026 is something the company calls AI Control Tower -- a governance layer built on top of its existing CMDB asset management infrastructure that lets enterprises discover, monitor, and manage every AI agent running across their organization. The metaphor both Zavery and Fipps kept returning to is air traffic control. "Imagine if you didn't have air traffic control and people were just flying around," Zavery said. "AI agents are not like humans. AI software can be very, very aggressive and very fast because there's no boundaries of their time or limits." Fipps described the commercial response as almost visceral. "I ask customers: how many agents do you have? Where are they in your organization? What do they have access to? Are they performing the way you envisioned?" he said. Most times, that conversation goes right to a need to see and engage with the AI Control Tower. He called customer uptake one of the biggest surprises of the week: "Pleasantly surprised" by how fast customers are engaging and wanting to contract for it. The real-world validation came from the customer panel. Melinda McKinley, COO of Strategy and Talent at Standard Chartered Bank, described scaling an AI assistant from a 50,000-person pilot in Hong Kong to 85,000 colleagues globally -- with case deflection rates climbing from 77% to 90%, triple the industry baseline. "AI is only as good as the data behind it," she said. "You have to be intentional about keeping that knowledge base live, current, and trusted." Oliver de Wilde, head of ServiceNow's Centre of Excellence at Hitachi Energy, described a 10-fold spike in employee self-service usage the week AI went live across 70,000 employees -- and a 25% reduction in calls to the IT service desk. The service desk manager called him that week in shock at the result and asked "what's happening?" he said. "They knew it was coming -- but they couldn't believe the reduction they were actually seeing." Those saved hours, he added, became hard negotiating leverage in renegotiations with service providers. "When you can use it to renegotiate a contract, the savings become very tangible." Pressed on where we are in the AI buildout -- an industry parlor game that has consultants arguing over whether we're in the second inning or the fifth -- Zavery declined to commit to a number but said it could be any of the first three. "It's definitely nowhere in the middle," he said. "I think it's still very early days." The technology remains probabilistic and not always backward compatible. The societal and regulatory frameworks are still forming. The cost structures haven't normalized. Fipps framed the next phase in terms of his own family history. His father was a turbine mechanic who spent his career being lowered onto high-voltage lines to fix massive generators. "I think the future infrastructure buildout -- for our country, but mostly globally -- is going to be a renaissance around innovation and opportunity and GDP growth," he said. "At the power core, the infrastructure core, it's going to be so much fun. Because we're going to do it in such a different way." For ServiceNow, that means the grinding, invisible work: security, compliance, backward compatibility, governance across regulatory regimes that differ by country, industry, and agency. "Enterprise software was never sexy," Zavery told Fortune, citing his three decades of working in the space and what a contrast the recent AI boom has been. "The amount of time people building software in this space spend -- not just building features, but making it secured, compliant, guaranteed performance ... all those things are never sexy jobs. They're very heavy, painful, getting into the nitty-gritty, making sure you're solving the difficult problems. And when the user is using it, they would never see any of this stuff. It's all the work you have to do underneath the covers." For a $96 billion company whose entire value proposition is being the infrastructure layer that enterprises trust most, it's not a problem that this work is unsexy. It's the pitch.

ServiceNow Knowledge 2026 - the whole darn thing

As FOMO shifts to FOMU (Fear Of Messing Up) and hinders the widespread adoption of Artificial Intelligence (AI), ServiceNow is promising to be the AI hall monitor and security guard, trying to become the mandatory layer through which enterprise AI must pass. ServiceNow's framing - that enterprises are drowning in AI chaos - is not just marketing hyperbole. Most organizations have spent the past two years in a kind of AI race, buying and testing tools, spinning up pilots, building agents, and layering AI on top of existing software and workflows. Pressure from the board and the C-suite has resulted in a lot of experimentation, but limited enterprise deployments in production. What's holding AI back? Humans are worried about two things. First, that AI without human oversight will go rogue and do damage like deleting or leaking critical data or agreeing to damaging business terms. Second, that AI without controls will burn past LLM (Large Language Model) token budgets in the middle of the night and leave them with a huge bill and little to show for it. Both are legitimate concerns. Governance is an unglamorous word, but it is increasingly the one keeping enterprise technology executives awake. As AI agents proliferate, writing code, processing invoices, resolving IT tickets, interacting with customers, and making procurement decisions, the question of who is accountable for what they do becomes a legal, regulatory, and reputational issue, not just a technical one. ServiceNow's answer is Control Tower, a centralized control plane that tracks every agent, model, dataset, and identity across an organization's entire AI estate, regardless of which vendor built it. The new Model Context Protocol (MCP) Registry takes this further: it creates a private, vetted catalog of approved connections that AI agents can make to external services, enforcing at the infrastructure level the same standards an organization would apply to a human employee accessing a sensitive system. By positioning itself as the governance layer for AI built on competing platforms, ServiceNow becomes the controlling platform, even for customers that don't buy its AI products. Although ServiceNow Control Tower was first announced last year, enhancements and features unveiled at this year's conference were like security blankets being passed out around the room: ServiceNow also announced ServiceNow Otto, a new enterprise AI assistant; new AI specialists for customer relationship management (CRM), IT operations, Human Resources (HR), and security; an Action Fabric open integration layer; new data intelligence tools for discovery, lineage tracking, and real-time quality monitoring; and autonomous security and risk capabilities (based on the Armis and Veza acquisitions). For developers, ServiceNow is embedding its Build Agent into Cursor, Windsurf, Claude Code, and GitHub Copilot, so developers can build for the ServiceNow platform without learning ServiceNow. The 2026 Knowledge announcements put ServiceNow ahead of the game of most of its closest competitors in terms of capabilities for managing, monitoring, and ensuring enterprise AI doesn't go off script or off budget. They also put ServiceNow squarely in the business of managing and controlling the agents built on competitors' platforms - at least until the others catch up. This is particularly important as ServiceNow releases agents for autonomous CRM, HR, and other business functions. As customers look to make decisions about which agents to run to support which business functions, such as point providers, CRM, contact center, and HR software providers, or more generalized platforms with AI capabilities like ServiceNow, the platform governing those agents will have the upper hand in determining which agents do what - and which vendors get paid for them. With built-in performance, security, and value monitoring, ServiceNow will also have an easier case to make for its agents over competitors. ServiceNow is making a big bet that enterprises will pay a premium for a platform that makes AI governable, and that the complexity, risk, and FOMU of the current multi-vendor AI landscape will make that premium feel cheap. When the risks are high enough, enterprises do invest in governance and compliance. The question is how quickly AI agent risk reaches that threshold and if it happens quickly enough for ServiceNow to capture the governance market before competitors, including some it currently calls partners, build comparable capabilities.

ServiceNow wants to be 'AI agent of agents' with Otto platform and AI tools

SaaS giant ServiceNow want to move from being the 'platform of platforms' to the 'AI agent of agents' with launch of Otto and wider AI suite. The Las Vegas-based software giant used its annual Knowledge conference to unveil a raft of AI-driven products it says will help enterprises move from AI ambition to AI execution. ServiceNow, the New York-listed workflow automation company whose platform processes more than 100bn transactions annually for enterprise customers, has used its Knowledge 2026 conference in Las Vegas to make a sweeping series of AI announcements it is framing as the next step in business transformation. The centrepiece of the announcements is ServiceNow Otto, a new enterprise AI experience that brings together conversational AI, autonomous workflows, and enterprise search into a single interface. The aim is to let employees complete work end to end across systems without switching between tools. The broader platform push spans four areas the company calls 'sense, decide, act, and secure'. On the governance side, ServiceNow expanded its AI Control Tower, which now covers more than 30 enterprise integrations and offers real-time observability into agent behaviour, automated compliance controls, and financial dashboards for tracking AI spend. New identity governance capabilities, delivered in partnership with Veza, extend oversight to human, machine, and AI agent identities simultaneously. On the execution side, the company introduced a new generation of AI specialisms for CRM, IT operations, employee experience, and security. ServiceNow claims its own internal deployment of the technology already handles over 90pc of employee IT requests, with cases resolved 99pc faster compared to human agents. CEO Bill McDermott framed the announcements in ambitious terms, describing ServiceNow as moving from a platform of platforms to what he called "the AI agent of agents". The company is targeting more than $30bn in subscription revenues by 2030, with AI expected to account for over 30pc of annual contract value. Customers including Honeywell, PayPal, Booking.com and the NHL were cited as live deployments of the platform. Separately, ServiceNow announced that its University learning platform has grown to nearly two million learners in the year since launch, up 80pc year on year, as demand for AI skills certification accelerates. Don't miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic's digest of need-to-know sci-tech news.

ServiceNow bids to become the control tower for enterprise AI - SiliconANGLE

ServiceNow bids to become the control tower for enterprise AI ServiceNow Inc. today unveiled a broad expansion of its artificial intelligence platform, stressing governance, security and autonomous execution as foundational requirements for enterprise AI adoption. The announcements center on enhancements to the company's AI Control Tower (pictured), a governance hub that allows organizations to monitor, manage and secure AI models and agentic workflows. The company also introduced new autonomous security capabilities and an expanded "Autonomous Workforce" of AI agents designed to execute specific business processes. Executives framed the updates as a response to growing enterprise concerns that AI deployments are proliferating faster than organizations can manage or govern them. "Customers are telling us AI is everywhere, but it isn't connected, isn't governed and isn't finishing the work," said Nenshad Bardoliwalla, group vice president of product management at ServiceNow. "The AI Control Tower for business reinvention is our answer." Introduced in 2025, AI Control Tower has evolved from a monitoring tool into what ServiceNow describes as a centralized command system for enterprise AI. The platform now spans five core functions: discovery, governance, security, observability and financial measurement. The system is designed to identify AI assets across heterogeneous environments, including major cloud providers and enterprise applications, while applying policy controls across models, datasets, prompts and agents. A key addition is deeper runtime observability, allowing organizations to track how AI agents make decisions and intervene when necessary. The platform also integrates risk frameworks aligned with regulatory standards such as the National Institute of Standards and Technology's Cybersecurity Framework and the European Union's AI Act. ServiceNow is also introducing an AI Gateway to provide real-time controls over third-party AI systems, extending governance beyond its own platform. The goal, executives said, is to move beyond fragmented tooling toward a unified system capable of managing AI at enterprise scale. The expansion comes as organizations grapple with a surge in nonhuman identities and autonomous agents operating across information technology environments. "AI-powered adversaries are moving faster than teams can human respond to or even detect," said John Aisien, senior vice president of central product management at ServiceNow. To address that gap, ServiceNow is introducing Autonomous Security & Risk, which integrates capabilities from recent acquisitions Veza Inc. and Armis Inc. The combined offering provides visibility into both asset inventory and identity access relationships, enabling organizations to map who or what has access to specific systems and data. The platform correlates signals across assets, permissions and decision processes to identify risks such as unauthorized data access or policy violations. It can also trigger automated remediation workflows, with optional human approval. Executives emphasized that the approach reflects a shift from reactive security models toward continuous, AI-driven governance. ServiceNow is also expanding its Autonomous Workforce initiative, introducing AI "specialists" across IT, customer service, customer relationship management and risk management functions. The agents are designed to go beyond discrete tasks to execute complete workflows. Launched in February Autonomous Workforce is "an entirely new class of AI specialists that think, act and work as part of a team right alongside your people," Bardoliwalla said. The new specialists can handle activities such as incident resolution, service requests and sales workflows, operating within predefined governance frameworks. Because they run on the same platform, they share data context, workflow orchestration and policy controls. ServiceNow executives said the model addresses a key limitation of earlier AI deployments, which often provided insights without execution capability. Rolls-Royce Holdings PLC has realized 5,000 hours of efficiency savings in IT operations alone through the use of ServiceNow's Now Assist generative AI platform and is seeing broader productivity improvements across manufacturing workflows, said Phil Priest, head of global business services at the auto maker. The company's experience underscored the importance of data readiness and governance, since AI deployments can amplify existing inefficiencies if underlying data and processes are not properly structured, he said. "As we expand Assist beyond IT to other functions, we have to almost rewrite our knowledge articles to make them AI-ready," he said, "so when the agent deploys it, it's done in a way that humans can get the answers they need fast and right the first time. Automation still requires good data to get the right answer." Priest estimated AI initiatives have had their biggest impact in manufacturing. "We've delivered 300,000 saved shop floor hours," he said. "That's real money." ServiceNow's broader strategy is to consolidate AI capabilities into a single operational platform that integrates data, workflows and governance. Executives said point solutions and disconnected tools are insufficient for managing the scale and complexity of modern AI deployments, particularly as agent-based systems become more prevalent. By combining observability, identity governance and workflow automation, ServiceNow is positioning its platform as an enterprise control layer for AI-driven operations.

ServiceNow just unveiled an AI workforce that can run your entire company: 'Enterprises need AI that senses, decides, and securely acts' | Fortune

ServiceNow used its biggest annual stage to make one sweeping argument: the era of AI as a helper is over. The era of AI as a worker has begun. At Knowledge 2026, held this week at the Venetian Expo Center in Las Vegas, the enterprise software company -- valued at roughly $95 billion and increasingly positioning itself as the operating system of the AI-powered enterprise -- unveiled a wave of announcements designed to move AI from the margins of business operations to the center of them. Taken together, they represent what ServiceNow considers the most ambitious product moment in the company's history. The centerpiece announcement was a major expansion of ServiceNow's Autonomous Workforce: a suite of AI "specialists" that don't just assist human workers but complete entire business processes from start to finish, without human intervention. The new AI specialists span IT operations, customer relationship management, HR, finance, legal, procurement, as well as security and risk. Unlike task-based AI tools or chatbots, ServiceNow says these specialists are role-scoped, governed, and embedded in proven enterprise workflows -- meaning they can triage a security incident, resolve an employee HR case, or close a sales quote autonomously, while leaving a full audit trail behind. Early results include ServiceNow's internal AI specialist resolving IT service desk cases 99% faster than human agents. Docusign is targeting autonomous resolution of 90% of all IT tickets. Honeywell says its AI assistant has eliminated the majority of service desk conversations. The city of Raleigh reports a 98% deflection rate on employee requests, saving the equivalent of a full month of staff time. "Advisory AI has run its course," said Amit Zavery, ServiceNow's president and chief product officer. "Enterprises need AI that senses, decides, and securely acts." The scale of the existing platform gives the pitch real weight: 23 million employees use ServiceNow's employee portal every month, generating an estimated 40 million-plus cases annually. The company says AI specialists across its customer base already resolve 91% of cases without reassignment. Each month, its CRM platform resolves over 100 million customer cases and configures more than 7 million quotes. The more enterprises deploy AI agents, the more urgent a second problem becomes: nobody knows where all those agents are, what they're doing, or who approved them. ServiceNow's answer was introduced at Knowledge 2025: the AI Control Tower. This year, the company announced that all AI Control Tower capabilities are now included across every product and package on its platform, built in by default rather than sold as an add-on. The Control Tower continuously discovers AI agents as they appear, risk-scores them, enforces least-privilege access, and measures their business impact against governance standards. The company also deepened its partnership with Microsoft to extend AI Control Tower governance across the Microsoft Agent 365 ecosystem. The integration gives IT administrators visibility into AI agents operating across both ServiceNow and Microsoft environments -- regardless of where those agents were built -- and allows ServiceNow's AI specialists to operate inside Microsoft 365 tools like Outlook, Word, and PowerPoint with metered usage tracked across both platforms. "One of the most important things we can do for enterprises is bring intelligence and action together in a secure, connected way," said Charles Lamanna, Microsoft's EVP of Business Industry Copilot. ServiceNow's security and risk division crossed $1 billion in annual contract value last year -- one of the fastest-growing segments on its platform -- and the company is doubling down with a new product called Autonomous Security & Risk. The launch integrates two recent acquisitions: Armis, which delivers continuous asset intelligence across IT, operational technology, IoT, and connected devices; and Veza, which maps every human and non-human identity and permission across an enterprise environment in real time. The combination gives security teams -- for the first time -- a unified picture of what exists in their environment and who or what is permitted to interact with it. The business case is urgent. As companies deploy more AI agents, those agents multiply the number of non-human identities operating inside enterprise systems, each with access to data and the ability to take consequential actions. Most enterprises cannot answer basic questions about those identities, such as who approved that access, why it exists, and whether it remains valid. Early customer results are striking. ServiceNow said a global energy company operating across 70 countries cut threat containment time by 97% and saved 1.2 million hours by automating security operations. A major U.S. financial services institution eliminated 96% of dormant non-human identities. A Fortune 100 aerospace manufacturer reduced control attestation time by 75%. ServiceNow also announced an expanded partnership with Nvidia, integrating Nvidia's accelerated computing infrastructure with the ServiceNow AI Platform, a move designed to give enterprises faster, more efficient AI agent deployment at scale. On the workforce development front, ServiceNow University -- the company's free learning platform -- has grown to nearly 2 million learners, up 80% year over year since its launch at Knowledge 2025. Two new tools debuted: AI Learning Guide, a conversational coaching companion that builds personalized learning paths, and SimStudio, a hands-on simulation environment where employees practice real ServiceNow tasks before going live. The World Economic Forum projects a net gain of 78 million jobs by 2030, with AI and big data topping the list of fastest-growing skills, and ServiceNow is clearly angling to be the training ground for the workers who fill them. Finally, a deepened partnership with Lenovo integrates its real-time device intelligence platform with ServiceNow's workflows, enabling enterprises to resolve up to 40% of IT issues proactively -- before users even notice a problem -- while cutting IT support costs by as much as 30%. Taken together, the announcements reflect a company that has made a definitive strategic bet: that the enterprise of the near future runs on autonomous AI workflows, governed by a central platform, and that ServiceNow intends to be that platform. The bet is well-timed. Enterprises are moving fast -- sometimes faster than their security, compliance, and governance infrastructure can keep pace. ServiceNow's argument is that it uniquely solves both sides of that equation: deploying AI at scale and governing it at scale, on the same platform, with the same data. Whether customers agree -- and whether autonomous AI specialists deliver on their promise at the scale ServiceNow is projecting -- will become clearer as the year unfolds. The security and risk AI specialists don't hit general availability until September. The IT specialists arrive in June. For now, ServiceNow is betting its next decade on the idea that AI agents will be our colleagues.

ServiceNow's Nick Tzitzon - 'The AI mess will land with IT. We have that history and will solve this for the whole enterprise'

Nick Tzitzon is Vice Chairman at ServiceNow, a position that spans customer relationships, market strategy. Tzitzon has been in the enterprise software business for a long time and he's an executive that has the institutional memory that is valuable during times of disruption. I sat down with Tzitzon mid-week to get his read on everything from ServiceNow's competitive positioning to the vendor's acquisitions to buyers' internal challenge of workforce transformation. The week's announcements - an expanded AI Control Tower, an Autonomous Workforce extending into every major enterprise function, and the first full integration of Armis and Veza into the platform - gave us plenty to work through. What Tzitzon provided me with was a useful reminder to look at recent history to understand where the 'agentic AI chaos' is heading - namely that humans and organizations will likely make the same technology mistakes that they have made in past decades. And that the IT department will be the ones to figure out how to make agentic AI practical in an AI context. ServiceNow CEO Bill McDermott has been pointing to enterprise complexity as the core problem: 367 applications, AI bolted on as a sidecar to each of them, and a CFO who can't find the ROI. However, instead of getting caught in the weeds, which is easy to do as a buyer when so much is coming at you, Tzitzon frames the same problem through the lens of what always happens next. He said: Every time there has been a new wave of exciting, transformational technology in the enterprise, there has been a shuffle of the chairs around the office, a new special initiative, and then a bit of time passes and the CEO gets frustrated and points to the CIO and says: 'What are we doing here?' His argument is that agentic AI will follow the same pattern - and that ServiceNow is the one company positioned on the right side of that equation. The long-standing relationship with the IT department and the CIO is not incidental to the vendor's platform story. Tzitzon said: We need to make sure they're given the right set of technologies to do that job in this generation of solution sprawl. It just so happens it'll be AI agent solution sprawl this time - but it'll still be sprawl. This is the most sensible argument for why other platforms - Google Cloud, Salesforce, Microsoft - might not be able to credibly claim the same orchestration position at ServiceNow. ServiceNow is not necessarily saying it has better technology (depending on who you talk to), but what It is explicitly saying is this: the CIO has cleaned up every previous mess, will clean up this one, and has been doing it on ServiceNow's platform for two decades. Tzitzon said: My gut instinct is that the long-standing relationship with the CIO and the IT department will serve us well in this agenda. Coming at it from CRM or HR - trying to take the cross-enterprise view from there - feels structurally harder. The chaos is ServiceNow's opportunity, in other words. That east-to-west argument - that going cross-enterprise on AI governance is an easier journey from the IT function than from any other starting point - is one McDermott also pressed in his media sessions this week. The logic is that the IT department approaches enterprise technology with a security and governance instinct that departmental leads simply don't have. IT sets the standards, knows what's required, and mandates the approach. In the age of autonomous agents executing business processes, that matters considerably more than it did in the age of SaaS procurement. Tzitzon also wanted to highlight that the people who have historically been deep in ServiceNow's core platform - ITSM, IT operations, IT asset management - have tended to operate within a defined scope. An important but narrow scope. Tzitzon's view is that those same people are, right now, the most relevant people in their organizations - because they are the ones best placed to govern and regulate agentic AI at enterprise scale. He said: I would love to take that core group and say: "You don't realize it, but you are the most relevant person in your organization, because you can solve a problem that is building right now that no one in any of these other areas knows how to solve. On the acquisition strategy - which attracted considerable attention when Armis was announced at $7.75 billion, ServiceNow's largest deal to date - Tzitzon said: So much of the potential of agentic AI is that AI agents can actually do things. But you can't do things if all you can access is IT systems. He described what a truly autonomous enterprise requires as a complete field of connectivity - every asset, device, and operational endpoint brought into a single integrated view. Without it, he said, you are: ...building subway cars without the subway system. You'd have these super powerful, fast things that can transport a lot of people, but they can't go anywhere. Healthcare was his reference point for why this matters in practice - an environment where every organization has a sprawl of connected devices that nobody has properly mapped. Getting those assets connected is something Armis can do near-term. Building the full agentic infrastructure on top of it takes longer, but the point of the acquisition was not to miss the moment while building organically. McDermott announced this week that ServiceNow plans to exit 2026 with the same headcount it entered with (a very different approach to other vendors in the industry), while simultaneously expanding its autonomous workforce. I asked Tzitzon what the company has learned internally about managing that kind of transition - and his answer was refreshingly honest: It's really hard, to be honest. The difficulty, he said, is not technological. It is cultural. People have been told repeatedly across their careers that a great unlock is coming - and they are weathered by it. He added: So when they hear, "Because of agentic AI, you'll be able to do far more with less," the first reaction, even in a tech company, is scepticism. Because everybody's heard it before. Trying to get people to feel comfortable saying "I shouldn't have to do that" is hard. Nobody really leans in culturally and says: "A significant portion of how I spend my day is work I shouldn't be doing." But that's actually the thing. It's really a cultural challenge before it's a technological one. He also made an observation about how organizations - including ServiceNow itself - tend to fall short at the next step. Telling someone that 90 per cent of their previous tasks are now resolved by AI is not enough if you haven't also told them what that opens up for their career and their contribution. His point, drawing on a comment attributed to Nvidia CEO Jensen Huang, was that the conversation leaders need to be having is not about what people no longer have to do - it is about what they can now do because of it. On governance frameworks - something we are hearing consistently from our CIO network is underdeveloped in their organization - Tzitzon's view is that the platform gets organizations "a solid three-quarters of the way there." The remaining quarter involves judgements every organization will have to make for itself, rooted in its own values. The analogy he used was schools deciding where the line sits between acceptable AI use and academic dishonesty. You can legislate broadly, but interpretation will always be local. diginomica noted from ServiceNow's financial analyst day at the start of the week that ServiceNow plans to double its revenues to $30 billion in just four short years. On that aggressive growth path, the argument is about addressable market as much as it is about product. Tzitzon noted that ServiceNow's customer count - around 8,000 when he last had direct visibility of the figure - sits against SAP's approximately 480,000. Tzitzon said: There is a substantial amount of real estate in the enterprise that ServiceNow has not historically reached, primarily because of how long we spent as the system of record for IT. Our home base was big enterprises. Well, now you have EmployeeWorks. You have CRM. You have data products. You have security. These things can not only reach the broader enterprise, they can reach the upper end of mid-market and, in some cases, full mid-market. And you also have an ecosystem that is far more versatile in seeing ServiceNow as a strategic platform - which historically has not been the case. ServiceNow's approach, as I consistently find it to be, is to ensure that its the value exchange with its customers is pretty equal. Making a subtle suggestion about some pricing tactics elsewhere in the market,Tzitzon's framing was: Being able to be the grown-up in the room when it comes to commercializing AI is going to help us And on what success actually looks like - beyond the $30 billion subscription revenue target - Tzitzon stripped it back to something more fundamental. ServiceNow had just come from a meeting with the global board of directors of its user groups - people, he said, who are "very opinionated about where we need to improve." His read was: The only success metric that genuinely matters is: are you moving the needle in terms of customers seeing real progress? Renewal rates, net promoter scores, and customer-facing metrics follow from that. The financial metrics follow from those. Everything else, in his framing, is downstream of whether the customer relationship stays intact and the platform keeps delivering. As I have written across this week's coverage, ServiceNow is moving fast and with intent - but the evidence of whether that translates to customer value is still accumulating. The conversations on the ground at Knowledge 2026 suggest momentum is real - and the customer stories have been telling in terms of the appetite from buyers. My lasting impression after the conversation with Tzitzon is that ServiceNow has invested over the past year to ensure it has a platform that is ready to capture the inevitable mess that emerges from over zealous AI adoption by buyers, convinced by the promise emerging from players elsewhere in the market. And when that mess lands in the hands of IT, whenever that may be, that's going to be a gift to ServiceNow.

ServiceNow Knowledge 2026 - beyond pilots: EMEA customers on AI at scale

ServiceNow opened Knowledge 2026 in Las Vegas this week with an expanded AI Control Tower and a push to bring its Autonomous Workforce into every major enterprise function. Big platform announcements are one thing, but customer validation is always what diginomica seeks to secure. With that in mind, this week at Knowledge 2026 we got to hear what ServiceNow EMEA customers are actually running in production. Cathy Mauzaize, ServiceNow's EMEA President, moderated a customer panel that brought together Her Excellency Nadia Ali AlThaibani, Executive Director of Common Digital Platforms at the UAE's Government Enablement department (DGE); Jayant Deulgaonkar, Head of SIAM Technology at Siemens AG IT; Mark Gerrard Douglas, VP Digital Services at SLB; and Christian Metzner, Managing Director HR & IT at Dirk Rossmann GmbH. The customers brought with them some really useful talking points for adopting automation and AI at scale - with numbers that back up the vendor platform story. Most of which certainly do not look like pilot projects. To give you an idea of what we are talking about: Siemens processes 1.7 million IT requests per year, with 96 per cent handled automatically; Rossmann reduced headquarters headcount by more than 50 per cent in five months, deploying six AI agents across classification, prioritization and incident resolution; the UAE government moved self-service adoption from 1 per cent to 54 per cent across more than 50 government entities, with 93 per cent of those requests closing within SLA; and SLB took external customer satisfaction from around 80 per cent to 93 per cent after deploying ServiceNow to its petrotechnical expert support function. Deulgaonkar set out how Siemens structures its operating approach: Our approach is: first, avoid any incident in the first place - predictive. If that doesn't work, prescriptive - meaning the system is self-healing and an agent can resolve it autonomously. Only if that doesn't work does a human agent come into the picture. Siemens is already at 92 per cent automation. Its Vision 2030 strategy is built around two pillars - "touchless" and "effortless support" - with AI now reaching into resolution problems that conventional automation could never handle. Deulgaonkar said: With AI we can now do things that conventional automation could not achieve - decision-making, complex resolution. The automation figures raise an obvious question about people. The panel's answers varied considerably on this point, with some speaking positively about reduced workforce requirements, whilst others pointed to the opportunity to move employees into higher value functions. At Rossmann, the 50 per cent headquarters reduction might sound jarring, but Metzner's argument is that headquarters exists to serve the stores, and had not been doing so well enough. The company operates more than 5,200 stores across nine countries, with roughly two million customers walking through German stores daily. He described forcing his headquarters team to work a day in a store: I forced - and 'forced' is a harsh word, but I did - my people to work a day in a store, so they really got the message about why we're doing this. At SLB, the position is different. Gerrard Douglas said: My priority is explicitly not to get rid of anyone, because geophysicists and geologists are a very rare commodity. The intent is to free them from more mundane tasks like filling in tickets, so that we can run more consulting engagements with customers, generate more revenue, and make growth. He cited an IKEA example - where helpdesk staff transitioned into interior designers, reportedly generating $1.3 billion in additional revenue - as the model SLB is working towards. SLB has been doing AI in various forms since 1980, but the ServiceNow layer is about redirecting rare human capital, not removing it. One of SLB's most helpful examples was what Gerrard Douglas called "carry context" - an approach to customer-facing support designed to eliminate the friction of explaining who you are and what problem you have. ServiceNow Engagement Messenger is embedded into every interface in SLB's applications, feeding user context from the control plane directly into ServiceNow at the point of engagement. A geologist in an oil company in Alaska arrives in the right specialist queue immediately - no password, no location, no explanation of which product they are using. A few years before, we had 7,000 tickets going to a generic service pool. They now go to the right person, first time. He added: If you go to a pump in the middle of an oil field and scan a QR code, we should already know what that pump is and which company it belongs to, and that should take you directly to the right instructions. Rossmann has built the same logic into its store interface. A colleague can raise a ticket by taking a photograph, add a message in any format or language, and pull product information via a barcode integration with the point-of-sale system. Metzner said: Every second matters in a store - it's real money. The change management challenge came up consistently across the session - as it has throughout Knowledge 2026. At Rossmann, Metzner noted the amount of investment required on this front: Looking at the whole project, at least 20 to 25 per cent of the entire effort went into organizational change. That included a dedicated sub-stream, an internal team, an external partner, town halls, daily communications and gamification alongside each other. The underlying challenge at Rossmann is the cultural distance between headquarters and the store floor - and the technology has to be invisible enough to bring those two entities closer together. As we all know, data prerequisites and getting your enterprise data house in order to see the benefits of automation and AI are some of the key challenges in getting results. Deulgaonkar said that Siemens spent two years standardizing its platform and cleaning its CMDB before AI could be realized at scale: The CMDB is critical - make sure it's really well set up from the beginning. Metzner pushed on this common assumption a bit, however. Rossmann went live in 20 stores after eight months of implementation, without resolving its data issues first: What we said we would absolutely not do was a two-year project cleaning up all the data before going live. We'd find out what data actually needed to be clean once we were live and learning. It depends on your risk appetite and your starting point - but I don't want to imply data doesn't matter. It does. You just don't need to solve it all before you start. That's probably a different approach to many here - and it was one of the fastest implementations we know of. You learn faster from doing. The company is now live across 800 stores, deploying into the next 50. It is worth noting that Rossmann is family-owned with no quarterly earnings pressure - a different risk appetite to most enterprises in the room. But as a corrective to the standard advice, the results are hard to argue with. A question from the audience on commercial leverage produced a degree of candour that is not always easy to get on a vendor event panel - a number of the customers spoke about the challenges of lock in, the price risk they are facing and how they work on their partnership with ServiceNow. On whether it is actually realistic to walk away from a deeply embedded platform, Gerrard Douglas - a former head of commercial software sales - noted that it is not easy, it costs money, but that it rarely reaches that point because the vendor has more to lose from the relationship ending. He said: You go find the leverage, have the tough discussion, and hopefully get to the right place without having to go through all that pain. Metzner described negotiating a co-investment arrangement with ServiceNow at the outset of the Rossmann project: I said to ServiceNow: we want to grow, you want to grow. What is your share of risk here? What are you willing to bring to the party? We found a very good agreement where ServiceNow invested in our project. AlThaibani framed the public sector version somewhat differently. The UAE's partnership includes a ServiceNow commitment to build a sovereign cloud, with DGE positioned to be the first government to benefit from it: It's not just about price - it's about how we ensure that the services we provide are secure in our own cloud environment. Summing up the overall sentiment of the session this week though, Deulgaonkar spoke about how organizations need to keep focused on the standards being set - rather than each new product release. He said:

Partners Tout ServiceNow's Innovation Engine: 'Beginning To Unlock The Age Of AI'

'I think [Service Now's expanded capabilities are] going to bring a level of AI capability down to the lowest-level operators and users in the platform, and in doing so, it's going to foster more innovation and more acceptance of these things. We see this as really beginning to unlock the age of AI, and I think it serves the bigger missions that we're seeking to accomplish,' says Jon Reynolds, co-founder and CEO of ServiceNow partner Naitiv. ServiceNow Tuesday opened its Knowledge 2026 conference in Las Vegas with a massive blitz of new technologies aimed at showing that it is a leader in bringing AI to the enterprise. The star of the show was ServiceNow's new Australia release of its Now platform, which builds on multiple innovations the company introduced in early 2026, said Nenshad Bardoliwalla, group vice president of AI. The innovations include Autonomous Workforce, which offers AI agents that think, act and work as part of a team along with human employees; added technology from its Moveworks acquisition, which develops front-end AI assistant and conversational enterprise search capabilities across all of an enterprise's systems; integrating Moveworks into ServiceNow's EmployeeWorks ahead of schedule; Context Engine, which shows AI who owns what to help determine actions; and Build Agent skills to let developers build AI agents with any development tools they prefer and then deploy and govern them, Bardoliwalla said in a pre-show press conference. [Related: ServiceNow CEO: We Are 'On Track For Our Best Year Ever'] "Together with a new commercial model that bundles everything customers need to deploy AI quickly, we've made it clear: The era of 'sidecar AI' is over," he said. What ServiceNow is doing in terms of the expanded capabilities it is bringing to its road map and the infusion of AI now across every SKU is exciting, said Jon Reynolds, co-founder and CEO of Naitiv, a Denver-based solution provider and ServiceNow channel partner. "I think it's going to bring a level of AI capability down to the lowest-level operators and users in the platform, and in doing so, it's going to foster more innovation and more acceptance of these things," Reynolds told CRN. "We see this is really beginning to unlock the age of AI, and I think it serves the bigger missions that we're seeking to accomplish." Enterprises are saying that AI is everywhere, but it isn't connected, governed or finishing the work, Bardoliwalla said. ServiceNow is looking to fix that with a significant upgrade to its AI Control Tower, which was introduced at last year's Knowledge conference. Bardoliwalla said the new AI Control Tower has evolved from a focus on visibility and management into a more comprehensive end-to-end offering with four new integrated capabilities: The new AI Control Tower works across five different dimensions, Bardoliwalla said. AI Control Tower gives ServiceNow a commanding position in the marketplace, Reynolds said. "I don't see another workflow platform at their level of prominence with the data model they hold or with the areas of business they impact," he said. "This really is their calling card. It's something that allows ServiceNow to be the 'control tower' of the enterprise." AI Control Tower represents the ability to provide control, orchestration and visibility to AI working across the enterprise while ensuring humans remain in the loop, Reynolds said. "That is what's going to unlock true autonomous AI where we can now begin to safely, ethically and responsibly remove humans from those loops," he said. Identity security is crucial for agentic AI where an application might have a primary agent that then generates subagents, and for a period of time the attack surface is expanded and then contracted, said Andrew Paolino, general manager for the U.S. for Konversational, a Dublin, Ireland-based solution provider and ServiceNow channel partner. . "What's cool about Veza is that when you pair that with AI Control Tower, you're telling a powerful story because now not only are you understanding where AI is in the environment and what it's doing the environment, but you're actually erecting guardrails around the AI as well," Paolino told CRN. "To me, that's a pretty interesting pairing in those use cases." ServiceNow is shifting to meet customers' evolving requirements around agentic AI and agents that need to be managed, said Jason Rosenfeld, chief growth and alliances officer at San Diego-based solution provider and ServiceNow partner NewRocket. "The AI Control Tower component is going to be really important," Rosenfeld told CRN. "There's a lot of messaging around AI Control Tower, and ultimately ServiceNow is positioning itself to be the orchestration and governance company for the enterprise, which is different from its historic focus on being the workflow company. Their messaging has completely shifted. I think it is the right messaging." The insight that AI Control Tower will provide is not only in the ServiceNow realm but across capabilities, said Jarred Pippy, COO of Everforth GlideFast, a Waltham, Mass.-based solution provider and ServiceNow channel partner. "We're potentially going to have customers on Salesforce using agents or Workday using agents," Pippy told CRN. "Having a product like AI Control Tower for governance and usage and getting insight is something we're going to pitch to our customers. We'll be happy to see that taking off in the near future." Bardoliwalla also introduced ServiceNow Otto, which unifies Moveworks and Now Assist into a single governed AI experience and entry point. "ServiceNow Otto is ServiceNow's new AI experience that turns intent into enterprise work for every person and across every workflow," he said. "Talk, chat, search, browse, analyze or build with ServiceNow Otto. You start it. ServiceNow Otto finishes it across every system and workflow involved on the platform that already runs your business." ServiceNow Otto started with Now Assist, which adds AI and GenAI into every workflow on the Now platform, Bardoliwalla said. "Then we added the AI experience as a unified front door with new modalities like voice, the Data Explorer, Lens and web agents," he said. "Moveworks brought the missing piece, a world-class conversational experience for employees. Otto unites all of it on a new AI-native architecture [with] truly agentic AI, multimodal interactions across every channel, and autonomous orchestration for complex work." Rosenfeld said NewRocket is already actively working with the Otto collaboration between Moveworks and Now Assist. "I think it's incredibly important for clarity for customers and partners to understand how this all works together," he said. "We've been a Moveworks partner prior to the acquisition, and the question was, 'How is this all going to work together?' I think the front door will be Moveworks, the interface that customers work with, but Now Assist is doing all of the actual work by AI in the background and feeding the Moveworks agents. That clarity and how that's architected is really important for customers and for partners." ServiceNow's acquisition of Moveworks was a great move and will make ServiceNow Otto a very useful tool for AI, Pippy said. Moveworks allows users to, in a single frame, search across platforms, he said. "If you're a user, think of it as sitting in a ChatGPT screen, and it's like, 'Hey, I have questions on this deal,' and it can systematically go look in Salesforce and pull that information for you," he said. "Or if you need approval for a new computer, it will go into ServiceNow where those approvals are. 'Hey, I need to get a [purchase order],' and it can go look in Coupa or Fieldglass or some other tool to pull that information. That's what Moveworks' selling point is. It allows the user to stay in one place but be able to take action on everything from that single pane. That's the future of where AI is going. ServiceNow, with that acquisition, is in a very good place, and we're happy to be on that journey with ServiceNow." ServiceNow Otto is a complete modernization of the entire user interface and experience, bringing it much closer to how most people are now interacting with a lot of technology, particularly with conversational AI, to corral multiple chatbots from several AIs together in one seamless experience, Reynolds said. "An example use case is, 'Transform the workspace of a claims processor,'" he said. "Your average claims processor has eight to 10 tabs in use at any one time, upwards of 20 tabs, different portals, and they're bouncing between them. I'm able to take this same exact UI, same exact capability, and now create a modern experience for that operator. The operator can just chat and send off communications to their customers, receive files back, and upload them to portals, and conduct AI analysis without leaving the chat screen. It's a huge step forward in the UI space." For the billions of autonomous agents that need to access the ServiceNow platform in a quick, frictionless and secure way without the full sophistication, the company also introduced the ServiceNow Action Fabric, which Bardoliwalla said is a new way for any external AI agent such as Claude or Copilot or a customer's own agent to drive real governed enterprise actions on ServiceNow through its MCP Server. "Others let agents read and write data," he said. "We let agents execute governed work. Our flows, our playbooks, approvals, catalogs, the full system of action, and all through the AI Control Tower so actions are identity-verified, permission-scoped and fully auditable. What we're hearing consistently from customers is that most organizations have more AI in production than they've inventoried or accounted for, and that anxiety around control, security and trust isn't going away. It's getting louder. At the same time, AI agents are taking real actions, moving real money, and affecting real people. But most of those agents are running without a system that governs them. That's not agentic business. That's agentic chaos."

Knowledge 2026 - CEO Bill McDermott was playing chess last year. Now he says ServiceNow manages everyone else's board

Last year at Knowledge 2025, ServiceNow CEO Bill McDermott told me his company was "playing chess, not checkers." I took that to mean a game of strategic patience - ServiceNow thinking several moves ahead while competitors reacted to the every latest update in the generative AI boom. He also declared at the time, in a notably sharp shift of tone, that CEOs were "now demanding software losers." The implication was that ServiceNow's AI Control Tower would become the arbiter of which enterprise applications survive the age of agentic AI. Twelve months on, here at Knowledge 2026 in Las Vegas, McDermott has continued to escalate that claim considerably - and with ever increasing optimism. The chess metaphor is back - but the ambition behind it has grown: We're playing chess. If you're a line-of-business toolset, you're playing checkers. Both games are perfectly okay - but we're playing a bigger game. We manage everyone else's agents too. They can't manage our agents because they don't do what we do the way we do it. In other words, this isn't just about ServiceNow winning a single game of chess, then. It's about dominance and managing the board everyone else is playing on too. It's a bold position and after attending multiple sessions this week, I'd also argue that it is an increasingly coherent one - and the argument for why comes down to a single word McDermott returns to again and again throughout the week: governance. Before examining whether the claim holds, it helps to understand the problem McDermott is describing. McDermott pointed to the financial crisis of 2008, which he argues, set in motion a decentralization of IT investment that has left most enterprises in a structurally difficult position for the age of autonomous AI. CEOs told line-of-business leaders to fund their own technology from operating expenditure and each department optimized for itself. The result, accumulated over fifteen years, is what McDermott calls the current reality: When AI arrived, it walked into an enterprise that had the big on-premise ERP systems and then, on average, almost 400 applications. Some are annoying toolsets. Some are a SaaS for everything. The enterprise got extremely complicated, and AI needs a clear path to be fully activated. The consequences of that complexity are visible in how knowledge workers actually spend their day. McDermott said: Every worker on average opens seventeen tabs a day to do their job. That's AI chaos, not AI control. And yet - despite this - most enterprises are still at the very beginning of meaningful agentic adoption. McDermott cited a figure this week that reflects what we see consistently across CIOs in the diginomica network too: Right now, only six out of ten companies have actually started the agentic business motion, and only one out of those ten has actually impacted business processes with true agentic AI. McDermott's argument is: the models exist; the compute exists; the appetite exists at the CEO level. However, what is missing, in his view, is the governance layer that makes enterprises confident enough to let AI agents actually execute: You can't have a probabilistic solution for an enterprise. It has to be deterministic and it has to be right every time. That's really important. This is something diginomica has been pointing out for some time - that enterprises aren't going to let AI tools run riot with their data, unless they can ensure deterministic outcomes. Generative AI is, by nature, probabilistic -- ask the same question three ways and you will get three different answers, McDermott said. That is interesting for consumer applications, but it is not acceptable for payroll, procurement, compliance, or order management. As McDermott puts it: When you're running a company and you want digital agents to work alongside humans - or even do the work humans are doing - they just have to execute along the lines of the business process so things actually get done. He is also clear about the risks of getting this wrong: "here are things that AI can do for your enterprise - but there are also things that AI can do to your enterprise. Doing things to your enterprise includes not having rules and rails where an agent can take down a company in nine seconds. That's not cool. That's dangerous. The governance proposition is also where ServiceNow's recent M&A strategy starts to make sense. When the company announced its acquisition ofVeza, and then Armis, I noted at the time that many observers - myself included - were uncertain about the sudden appetite for inorganic growth from a company that had historically grown organically. Arriving at Knowledge 2026 and seeing these acquisitions reflected in the product portfolio, the logic is considerably clearer. These aren't about acquiring growth in the way that SaaS vendors historically acquired companies to boost revenue. It's clear to me now that these are technology tuck-ins to bolster the ServiceNow platform in the area it's making its key argument: again, governance. Veza brings identity governance - the ability to track and manage what humans and AI agents can access across every system in the estate. Armis extends that visibility into operational technology, medical devices, and cyber-physical environments that most enterprises cannot currently map. Together they give ServiceNow the capability to credibly claim it can govern agents across the full enterprise surface area - digital and physical. On the identity challenge, McDermott said: Identities are more important than ever. Humans have a profile, rights, and privileges - as we all do when we work for a company. Agents have to be the same way. And the context argument - grounding AI decisions in enterprise knowledge rather than raw model output - is where he believes the real differentiation lies: We bring the context of 100 billion workflows and seven trillion transactions. When you meet that with the compute power of a language model, you have something that genuinely makes people more productive. I covered the Q1 2026 earnings in April, where I noted that solving the governance bottleneck was beginning to look like a genuine commercial differentiator. This governance piece also extends into ServiceNow's acquisition of Moveworks - which it snapped up last year and is now expressed as Otto, the unified agentic front door announced this week. Otto is doing different work to Veza and Armis but serves the same overall argument. Otto puts conversational AI in the hands of every employee, reaching across whatever systems sit behind it. The demos I saw this week were genuinely impressive: an employee changes their benefits, approves a hardware request, provisions system access for a team member - all within a single conversational interface that reaches into any underlying systems simultaneously. ServiceNow faces competition here, however, as Microsoft's Copilot and Slack's AI experience are both pursuing the same front-door positioning with significant distribution advantages. But the argument ServiceNow is making is that a conversational front door is only as useful as the governance sitting behind it. A front door that opens into chaos - seventeen tabs of chaos, in McDermott's framing - does not solve an enterprise problem. One that connects to structured workflows, identity management, and a governed control plane is a different proposition. I put the harder version of this to McDermott in the media session -- if ServiceNow wants to govern AI east to west across the enterprise, how does it facilitate what is historically a fragmented buying decision, with different departmental budgets, agendas and needs? His answer was interesting: The C-suite - specifically the CEO - is sending a clear signal. The CEOs of these companies are very intimately aware of their AI imperative, and they are involved in the decision-making process. He argues that the lesson of 2008 - where decentralisation of IT spend led to fragmentation - is something CEOs are now actively trying to avoid repeating. And his response to the competitive question - why ServiceNow rather than Salesforce or Workday - is this: What if there's a compensation issue on aisle seven in the finance department that needs to be resolved across legal, finance, HR, and the general ledger? Only one company can do that. It has nothing to do with market share debates - it's a bigger story. It's a different technology. It's a different system. And: We are the AI agent of the agents. For companies that are truly going to agentify their business, somebody has to be responsible for all those agents. He is also clear that openness is part of the trust argument, not just a commercial positioning choice: Totally open is our model. You see others moving in the opposite direction. When you close your platform, in my view, that signals a lack of confidence in it. As I wrote following the platform packaging announcements in April, ServiceNow's structural advantage in this conversation is the CIO and CISO relationship. The IT function approaches enterprise technology with a security and governance instinct that most departmental leads simply don't have - and that instinct becomes considerably more important when the technology in question is autonomous agents executing business processes. This is the conversation ServiceNow has been having with customers for two decades. The vocabulary has changed, but the buyer is the same. And now ServiceNow has increased security and governance capabilities to back it up. On the financial ambition, McDermott is committing to $30 billion in subscription revenues by 2030 - roughly doubling the current run rate in a few short years. He points to a remaining performance obligation of nearly $30 billion as evidence that the company is already tracking towards it: If your RPO is double your revenue, you're a healthy company. But the $30 billion commitment, he made clear at the Financial Analyst Day held the day before Knowledge opened, is not actually the number he wanted to give. The management team, bottom up, signed off on $30 billion. McDermott's own view runs higher. As he said at the analyst session: this is not the "Bill ambitious number" - that one didn't make it past the review process. He was similarly pointed about the upside scenario of $32 billion that Gina Mastantuono, CFO, outlined - and he said: Nobody here is talking about a ceiling. As I wrote earlier this week, I was heading into Knowledge 2026 with the anticipation that ServiceNow is entering its third era, having undergone a second reinvention. The argument McDermott is making is more coherent at Knowledge 2026 than it was twelve months ago - and that's in large part because the product portfolio now backs it up. The acquisitions that looked aggressive when they were announced fit into a governance story that is clearly outlined at this point: Veza for identity, Armis for cyber-physical visibility, Moveworks for the agentic front door. But there are real tensions here that McDermott's confidence sometimes masks. ServiceNow is moving very fast - deliberately, aggressively fast - in a market where enterprises are still slow to adapt. We are hearing from the diginomica network that buyers are not yet seeing the AI value they expected from deployments already underway - often due to the organizational and workforce implications of adopting AI.. In many cases the technology is running well ahead of the enterprise's readiness to absorb it. McDermott's response to that is the commitments he announced this week: a total satisfaction guarantee, a go-live promise in under 100 days, AI Control Tower free for a year. As I noted in the keynote coverage yesterday, this is competitive grandstanding as much as it is customer value - what went unsaid being the pointed question of why other vendors at this scale aren't making the same commitments. But there is also a straightforward commercial logic to the pace - that McDermott himself acknowledged. He wants to capture wallet share now, prove value early, build platform dependency before competitors establish a foothold. He said: "We are using this disconnect between Wall Street and Main Street as our opportunity to build, build, build and get ready for growth. Last year the question was whether ServiceNow had earned the right to claim the orchestration layer. This year the claim is more expansive: that governance is the thing enterprises are actually missing, that ServiceNow is the only platform that can credibly deliver it across the full enterprise surface area, and that the models, the hyperscalers, and even the competing agents are welcome - as long as they come through the control tower. Whether the enterprise market lands there is genuinely uncertain. The incumbents are not standing still and the hyperscalers have leverage nobody should underestimate. But the governance argument - deterministic, predictable, auditable AI that is "right every time" - is the right argument and one, I believe, will land very well with buyers. If ServiceNow can prove it at scale, McDermott will have earned considerably more than checkmate.

ServiceNow's AI Control Tower: Full-Stack Governance for the Enterprise

AI-native experiences, voice, and intelligent approvals show customers where AI drives the most value and embeds directly into the flow of work Today, at ServiceNow's annual customer and partner event, Knowledge 2026, ServiceNow, the AI control tower for business reinvention, expanded its AI Control Tower offering with new capabilities that give enterprises control over every AI system, agent, and workflow, regardless of where it runs. AI Control Tower, first introduced at Knowledge 2025, has evolved from visibility and management into a comprehensive, end-to-end solution that lets customers act with confidence across five dimensions: * Discover finds AI assets once deployed across the organization -- including systems beyond ServiceNow -- through 30 new enterprise integrations spanning Amazon Web Services (AWS), Google Cloud, and Microsoft Azure, and enterprise applications such as SAP, Oracle, and Workday. Discovery also extends to non-human identities and connected devices, bringing OT and IoT assets into the same governance model as AI agents and cloud services. * Observe provides continuous monitoring with live metrics and alerts, replacing periodic audits for ROI analysis. Through the recently completed acquisition of Traceloop, AI Control Tower now delivers deep observability into AI agent behavior at runtime, giving teams visibility into how agents reason, where they make decisions, and when to course-correct. * Govern delivers AI-driven risk assessment across all types of AI, not only agents but also models, data sets, prompts, and classic machine-learning. Five new risk frameworks aligned to NIST and EU AI Act standards provide compliance controls out of the box. * Secure extends identity access governance to hyperscaler AI environments and every connected device through integration with Veza, bringing patented access graph technology, scoped permissions, and least-privilege enforcement to every AI system, agent, and identity. When an agent goes off script or operates beyond its permissions, AI Control Tower can detect it and shut it down in real time -- giving organizations the kill switch they need as agents take on more critical work. * Measure provides cost tracking and ROI dashboards that give customers financial control as they scale AI -- addressing runaway model spend -- one of the most pressing challenges enterprises face as AI deployments grow. This approach is powered by two decades of enterprise operational data accumulated through 100 billion workflows and 7 trillion workflow transactions annually, and anchored by the ServiceNow CMDB and Context Engine, which is designed to map digital assets to the services, people, and processes it supports. Enterprises can sense signals across their full digital estate, decide with live business context, act through autonomous workflows, and secure every agent action. Standalone governance tools simply cannot replicate this. In addition to its existing integrations with Anthropic and OpenAI, ServiceNow has also announced deepened AI Control Tower integrations with AWS, Microsoft, NVIDIA, and other LLM providers, extending governance and observability across the infrastructure enterprises rely on most. For example, the ServiceNow AI Control Tower now integrates with the NVIDIA Enterprise AI Factory validated design for agent observability, extending governance and risk controls to the infrastructure layer of large-scale AI deployments. "Enterprises are under real pressure to deploy AI and show results, but there's a major gap between adoption and accountability," said Jon Sigler, executive vice president and general manager of AI Platform at ServiceNow. "ServiceNow AI Control Tower was built for this moment: delivering unified governance across the entire enterprise AI stack, so security and control move at the speed of the business." Additionally, for all customer Model Context Protocol (MCP) transactions, a new AI Gateway provides real-time controls for agentic workloads, with governance, observability, and security for full visibility across any third-party AI system. From AI investment to AI outcomes For many enterprises, the bottleneck is deciding where to apply AI first, validating that it works, and reducing the effort to deploy. New AI agents and innovations from ServiceNow form a connected loop that helps accelerate time to value from day one: * AI Agent Advisor analyzes each customer's operational data, including incidents, cases, and conversations, to identify the automation opportunities that will have the greatest impact in their environment. It identifies patterns in real workflows and matches them against available agents or helps create new ones. * AI-powered setup applies AI to the implementation process itself. Instead of time spent on manual configuration, AI agents can handle plugin installation, role provisioning, and system setup autonomously. For new customers, applications are ready to go the moment their instance is provisioned. * AI-powered center brings all AI administration into one centralized hub for setup, configuration, optimization, and ongoing management. The Evaluation Suite lets organizations validate that deployed AI is performing as intended, both before and during production. More than 150 customers have already used it across approximately 1 million AI interactions. AI in the natural flow of work Too often, enterprise AI lives in a separate window, a chatbot sidebar, or a tool that requires users to leave what they're doing to get the support they need. Users want to speak and ask questions in natural language and receive contextual answers while staying in their workflow. The ServiceNow AI Platform embeds intelligence directly into the surfaces, interactions, and modalities where work happens. The ServiceNow AI Platform delivers these capabilities across modalities, including voice, vision, and natural language, so employees experience AI as a seamless part of how they already work: * Device cameras populate forms and trigger actions from real-world context * Users get real-time, step-by-step help using voice and natural language across the platform with Dynamic Guidance. * Complex screens are transformed into clear summaries, improving accessibility and boosting productivity with Screen Summarization. * Employees can consume policy updates, runbooks, and training content on the go with SmartDocs and read policies in plain language, evaluate incoming requests in real time, and handle routine decisions instantly with Intelligent Approvals. * Key partnerships for voice AI give organizations the flexibility to use AI-powered voice with their existing contact center infrastructure including Amazon Connect, NiCE, Five9, 3CLogic, and Twilio. What customers say about the ServiceNow AI Platform Rolls-Royce "For more than a century, Rolls‑Royce has engineered the extraordinary, and now we're applying that same standard to how we work. ServiceNow AI has transformed our digital self‑service, delivering critical information directly into workflows. Adoption has nearly tripled, with 38,000 tickets deflected in a year and resolution times reduced by 34%," said Rachel Cameron, VP of performance & improvement at Rolls Royce. "We're now scaling that success across Global Business Services, introducing autonomous actions across IT, HR, and Finance to reduce manual effort and drive measurable outcomes. ServiceNow is a critical enabler of our journey toward self‑reliant service management, responsibly and at scale." HDFC Bank "As India's largest private sector bank, we operate at a scale where AI governance isn't optional, it's foundational," said Ramesh Lakshminarayanan, group CIO at HDFC Bank. "We run ServiceNow AI across IT and risk, and AI Control Tower is the common governance layer across all of it, giving us the visibility to manage every AI use case and the confidence to scale." Rossmann "Retail is all about efficiency, is all about speed. Being able to use voice is an absolute game-changer -- because every second matters in a store," said Christian Metzner, managing director HR and IT at Rossmann. "With AI Voice Agents, Rossmann's store associates will be able to resolve issues hands-free, in their native language -- keeping their focus where it belongs: on the customer." National Hockey League "At the NHL, working smarter has always been the goal so we can stay focused on delivering extraordinary fan experiences," said John Frantzeskakis, SVP of technology, operations and digital transformation at the National Hockey League. "The ServiceNow AI Platform gives us the AI control tower we need to scale AI with confidence, transforming fragmented operations across 32 clubs and 1,300-plus games a season into connected, intelligent workflows. We're already seeing real productivity gains, and we've only scratched the surface of what AI can do for us." Academy Sports "ServiceNow is a strategic differentiator, our platform of intelligence. As we begin this journey, IT is the engine, but we are architecting a digital twin of our operating footprint where AI connects our assets to our people," said Amyn Gillani, VP core technology and AI transformation at Academy Sports. "This expands the value radius far beyond the tech stack, rippling into HR, Security, Supply Chain and more; ultimately revolutionizing Retail Operations by empowering our associates and providing a superior customer experience." Availability Features announced today and many more as part of the ServiceNow AI Platform Australia release are available on a rolling basis beginning April 2026. AI Agent Advisor and Intelligent Approvals are generally available in May 2026. AI Control Tower enhancements enter Innovation Lab in May with general availability expected in August 2026. Full details can be found in the ServiceNow Store.

ServiceNow Knowledge 2026 - Standard Chartered, Hitachi Energy and State of Hawaii on why governing AI agents matters more than building them

ServiceNow opened its annual Knowledge conference in Las Vegas this week with a pair of major platform announcements - an expanded AI Control Tower and a push with its Autonomous Workforce into every major enterprise function. But vendor strategy is one thing - what customers are actually experiencing is another. As part of the media and analyst programme at Knowledge, Paul Fipps, President, Global Customer Operations at ServiceNow, moderated a customer panel session that brought together Melinda McKinley, Chief Operating Officer at Standard Chartered Bank, Tom Ku, IT Chief Operating Officer for the State of Hawaii, and Oliver de Wilde, Head of the ServiceNow Centre of Excellence at Hitachi. Interestingly, the customers supported what ServiceNow is laying out as its priorities, in that they spoke less about agentic AI capability and more about governance - specifically, whether organizations can actually put AI into production safely, at scale. Fipps opened with two stories that outlined why governance is a top priority for enterprises, when it comes to agentic AI deployments. Firstly, he explained that a CTO at a large financial services company had built 30 production-grade agents and hadn't deployed a single one. Fipps recounted the exchange: I asked how many he had. He said: I've built 30 agents - they're production-grade and operationally ready. I asked what they had access to, whether they were actually performing. He kept deflecting. Eventually he said: listen, I haven't actually put them into production, because I can't answer any of those questions - and I'm in a regulated industry, so I certainly can't put them there if I can't answer them. So we talked about AI Control Tower from ServiceNow - a control plane where you can actually manage agents inside ServiceNow across all your systems of record and your own customer agents. The second story came from a healthcare life sciences CIO who had cancelled 900 pilots - not because they were failing, but because he couldn't govern them. Fipps relayed his words: I've now cancelled them all - not because they weren't working, not because they hadn't done some really important things, but because I couldn't govern them, I couldn't control them, and I'm not going to have a bunch of custom software running around my enterprise. Standard Chartered provided a clear illustration of what governance-led AI deployment looks like in a regulated institution, and McKinley explained how the bank has approached it. Her remit spans Group Strategy, HR, Supply Chain Management, Property and Brand Marketing alongside Group Transformation - the enabling functions, as she describes them. The bank piloted a virtual assistant across around 50,000 colleagues in India and Hong Kong, targeting case deflection - reducing the volume of requests routed to live advisers. The result was a 77 per cent case deflection rate. The deployment has since scaled to around 85,000 colleagues globally, with around 90 per cent first-contact resolution across knowledge management, search and case submission. Fipps noted that Standard Chartered's figure was well beyond what ServiceNow had been targeting in early implementations, where the benchmark was 25 to 30 per cent. Getting to 77 or 80 per cent, he said, was "genuinely fantastic" - and it has become the foundation on which Standard Chartered's first colleague-facing agentic use case is now built. Commenting on the importance of governance, McKinley said: As a heavily regulated business, we're building that out to define our governance pathways intentionally - ensuring really clear decision-making, clear accountability, and effective management of AI across the organization. Hitachi Energy joined ServiceNow's Lighthouse programme in 2024, giving it early access to capabilities in development - including what would become AI Control Tower. De Wilde noted that it was genuinely a two-way relationship - Hitachi Energy and ServiceNow helped shape the product it is now using. The rollout moved from a pilot of a few thousand users to an enterprise-wide deployment in March 2025, reaching close to 70,000 people within six months. The results were immediate. De Wilde said: When we turned it on, we saw almost a tenfold increase in users moving towards self-service, and a 25 per cent reduction in calls to the IT service desk. The service desk manager called me - they obviously knew something was happening, but they couldn't believe the volume of reduction they were seeing. Those hard productivity gains had also become a commercial lever in supplier negotiations. He said: When you go back to contract those providers, you can point to hard, tangible savings - not just soft productivity hours, but genuine commercial leverage. Change management is consistently showing up as a blocker to AI returns in the CIOs diginomica speaks to regularly - and customers this week at Knowledge 2026 were keen to highlight its significance. On how organizations actually get teams to adopt and sustain AI capabilities, McKinley described how Standard Chartered built its change management and user experience expertise around six or seven years ago - several years before any AI implementation began. The starting point was research. She said: Several years before we implemented any AI, we did extensive user experience research: going out to the organization to understand sentiment towards AI, appetite for it, familiarity with it both in and outside the workplace. You get very interesting data from that, including generational differences. The approach that followed was deliberately segmented - different strategies for different persona groups, delivered through nudges in day-to-day workflows, reskilling programmes, gamification and use-case sharing. She added: Having that intentional strategy, listening to the feedback, and adjusting along the way is really, really important. The knowledge management piece was equally important. A virtual assistant is only as good as the data underpinning it, and keeping that current requires clear processes for update triggers and review approvals. Without a strong knowledge foundation, trust erodes quickly. At Hitachi Energy, HR was not an early advocate. De Wilde said: Our HR teams were not the biggest fans of this when it started. The IT teams were more open. But HR is now our biggest advocate. The reason was sequencing. IT had a more harmonized environment and moved first. HR watched, used the learnings to standardize some of their own processes, and then applied the same approach themselves. That sequencing - process before automation - was one of the most consistently made points across the session. De Wilde said: AI is the cherry on top - fix the process, understand it, and AI makes it better. McKinley explained how Standard Chartered put this into practice for the onboarding agent. Rather than automating the existing workflow, the bank deconstructed it entirely. She said: We didn't look at it as 'layer AI over the existing process.' Instead, we deconstructed every task across our entire onboarding journey - asking what the pain points are, what feedback we'd had from hiring managers and from new joiners. We broke those tasks down into skills, matched each skill to either AI, agentic automation, or human delivery, and then reshaped what the remaining human roles look like. It was a mass-scale work redesign. Hitachi Energy's position on ServiceNow is the product of an organizational and technology rebuild following the company's acquisition by Hitachi from ABB in 2020. With the opportunity to select platforms from scratch, ServiceNow became effectively the ERP for the IT environment. When AI arrived, embedding agentic capability into an already-consolidated platform followed naturally. De Wilde said: It's really about understanding your processes, identifying the right things to automate, and sometimes asking whether you actually need to fix the fundamental process first before automating it. ServiceNow was always the choice for us, and we will continue the partnership for the years to come. For Standard Chartered, the answer came back to the flow of work. The bank had been working with ServiceNow well before AI was prevalent. Reducing friction across its journeys - using automation or AI to address pain points - was the natural extension of a platform already embedded in daily operations. For the State of Hawaii, the more immediate story was what an autonomous implementation actually looked like in practice. ServiceNow's approach compressed what would normally be a multi-year process into weeks - with workshops focused entirely on business process, no technology conversation, and UAT reached within four weeks. Ku said:

ServiceNow expands AI agent governance through deeper integration with Microsoft

ServiceNow today announced an expansion of its strategic partnership with Microsoft that brings order to the chaos of AI agent sprawl. The partnership includes a deepened product integration between ServiceNow AI Control Tower and Microsoft Agent 365, extending AI Control Tower's existing governance across Azure-backed Microsoft Foundry and Copilot Studio to Microsoft Agent 365's AI agent ecosystem. ServiceNow AI specialists will also be available in the Microsoft Agent 365 Marketplace, allowing the ServiceNow Autonomous Workforce to operate across the Microsoft 365 tools that employees use. Governing AI agents is critical, now more than ever, as enterprises accelerate their deployment of agentic technology across systems, teams, and tools at unprecedented speed. Realizing the full value of this investment requires a unified approach to governance, identity and permission management, and control that spans every platform across which AI agents operate. ServiceNow is working with partners like Microsoft to give enterprises the visibility to see AI agents, models, tools, and prompts in their environment, which gives them the control to govern them consistently and the interoperability to put them to work across the tools employees already use. "ServiceNow and Microsoft are helping organizations maximize value from every AI investment," said Jon Sigler, executive vice president and general manager, AI Platform at ServiceNow. "With this expanded integration, customers can securely apply governance across ServiceNow and Microsoft environments with integrated visibility and controls, while putting ServiceNow Autonomous Workforce to work across the Microsoft 365 environment. This is an example of what it means to put AI to work at enterprise scale, with the trust and interoperability that business transformation requires." "One of the most important things we can do for enterprises is bring intelligence and action together in a secure, connected way," said Charles Lamanna, executive vice president, Copilot, Agents, and Platform at Microsoft. "That's why we're excited to collaborate with ServiceNow, bringing their AI expertise into Microsoft 365 to add workflow intelligence on top of that secure foundation. Together, we're helping customers act on insights more quickly and drive meaningful outcomes across their business processes." Unified control for a multi-agent enterprise ServiceNow AI Control Tower already connects to Microsoft Foundry and Copilot Studio to discover AI assets, enforce governance policies, and drive consistent oversight. The new AI Control Tower integration with Microsoft Agent 365 expands these capabilities by extending visibility and governance insights across Microsoft Agent 365's AI agent ecosystem. This gives IT and operations teams enhanced visibility into agent activity across ServiceNow and Microsoft ecosystems, regardless of where these agents were built or deployed. Through the integration, ServiceNow AI Control Tower gives administrators the ability to review and approve ServiceNow AI specialists prior to submission to the Microsoft Agent 365 Marketplace, where Microsoft publishing and policy controls apply. This helps ensure that every ServiceNow AI specialist that interoperates with the Microsoft 365 environment has been vetted, permissioned, and authorized for deployment. ServiceNow AI specialists come to work across Microsoft 365 applications In the Microsoft Agent 365 Marketplace, a ServiceNow AI specialist will appear in the org chart as a digital employee with defined roles, permissions, and accountability. This allows the AI specialists to be able to take actions such as drafting a Word document, responding to email messages in Outlook, or acting on an assigned comment in PowerPoint, subject to Microsoft 365 permissions, identify, and admin policy controls, while consumption is tracked across both ServiceNow's and Microsoft's metered usage models. Years of innovation for one autonomous future Today's announcement builds on years of collaboration between ServiceNow and Microsoft on behalf of enterprise customers, spanning cloud infrastructure, productivity, and AI. This is the next step in that shared commitment to making enterprise AI governable, interoperable, and scalable. In addition to the integrations announced today and as part of their ongoing partnership, ServiceNow and Microsoft are expanding their go-to-market motion to include autonomous IT. This combines ServiceNow's AI specialists and workflow intelligence with Microsoft's cloud infrastructure and productivity ecosystem to deliver governed, autonomous IT operations to mutual customers at scale. The AI Control Tower and Microsoft Agent 365 integration is available in preview, and ServiceNow AI specialists will be available in Microsoft Agent 365 Marketplace later this year. Learn more about how ServiceNow is expanding its Autonomous Workforce of AI Specialists to major business functions at Knowledge 2026 here, and how it's introducing new AI Control Tower capabilities to further discover, observe, govern, secure, and measure AI deployed across the enterprise here.

Knowledge 2026 - CEO Bill McDermott says AI intelligence is commoditizing, but chaos is coming

ServiceNow today announced an expansion of its AI Control Tower and Autonomous Workforce products, where it outlined how the platform's AI execution and governance capabilities are extending into every business function. The company's acquisition strategy in recent months is now showing up comprehensively in the product portfolio - Moveworks for the 'agentic front door' and Veza and Armis for its security and risk agenda. And CEO Bill McDermott took to the stage this morning to deliver his first mainstage keynote of the week, where he sought to ground these products in terms that he hopes will resonate with C-suite buyers. Before he got to the products, McDermott outlined how he views the current macro context. Whilst some argue about job displacement due to AI, he offered a different position - that the global workforce is aging, birth rates are declining, and the world is facing a labour shortage of up to 50 million workers by 2030. He said that the agents and robots are helpfully coming online at this exact moment and his position is that these are the ideal partners to complement a shrinking human workforce. Equally, he added, the companies that figure out how to deploy them safely will be the ones that grow - and the ones that don't will be exposed. In addition, McDermott spoke about why buyers need to understand that AI intelligence is commoditizing - essentially that they shouldn't be viewing models as the future differentiator - and positioned governance and execution as the key to expanding agentic AI adoption safely. The core pitch was that enterprises have a 'blind spot' when it comes to AI: that uncontrolled agents operating without identity, audit trail or compliance posture represent a serious business risk. He referenced the case of PocketOS as an example - where the company said that an AI agent deleted production databases in a matter of seconds. He also pointed out that cyber crime is already the third-largest economy in the world, behind only the United States and China. The more agents you deploy without governance, the more surface you expose. He said: Intelligence without rules and rails is a dangerous blind spot. To illustrate the enterprise AI chaos problem, McDermott painted a picture of what the average enterprise often looks like - an articulation of what many organizations will be experiencing right now. 367 different applications, with AI bolted onto every one of them like a sidecar - none of them connected or governed. A CFO who approved the spend but can't find the ROI - and a token bill going up every month. Six out of ten companies are already using agentic AI - but only one in ten have built anything autonomous. And employees, he said, toggling between 17 tabs wondering whether AI was supposed to make their lives easier or harder. His summary of where that leaves most enterprises and why ServiceNow is positioned well to support was: The real competitive differentiator is the orchestration surrounding the models. I've been reporting on ServiceNow's governance argument across the Veza acquisition, the Autonomous Workforce launch in February and the platform restructuring in April. What McDermott aimed to do this morning was take these product details and focus them in on something that makes sense to a CEO. It's also worth noting how McDermott characterized ServiceNow's own evolution. He acknowledged the company has previously stood on the Knowledge stage and described itself as the "platform of platforms." That, he said, no longer captures what ServiceNow is and that the updated pitch is that ServiceNow is now "the AI of agents". I'll be returning to this in future analysis, as I think it's worth exploring further. Openness and integration have been core to ServiceNow's philosophy for many years, but I think the challenge this time around is that there are others in the market making the same argument aggressively and there will be an internal struggle in enterprises about which platform, or multiple platforms, win out. The benefit ServiceNow has is its long-standing relationships with IT departments and CIOs. McDermott also sought to ground ServiceNow's execution credentials in the company's own internal data. The company runs what it calls Now on Now - ServiceNow, on ServiceNow. McDermott said that in 2025 the company saved half a million dollars on Now Assist alone within that programme, where 91 per cent of its own service requests are now supported by AI. And interestingly the event itself - 25,000 attendees, every incident in the venue, every service request - was running live on the platform. He said: You don't have to imagine what an Agentic business looks like. You are actually inside one. One of the central problems the AI Control Tower is designed to solve is the inability of most enterprises to measure what their AI investment is actually worth. The fact that ServiceNow can point to its own books, and to a live event as a proof of concept, is likely the kind of evidence that will resonate well with C-suite buyers. Our own diginomica network of CIOs have told us time and time again that they are struggling to get value from their current AI deployments. To bring the customer perspective, McDermott brought Raj Subramaniam, CEO of FedEx Corporation, and Vishal Talwar, EVP and Chief Digital and Information Officer of FedEx Corporation and President of FedEx Dataworks, on stage for a conversation about what responsible AI deployment looks like across a network moving 18 million packages a day. Subramaniam's said: The difference between business and technology has kind of evacuated...the role of running business is technology, and technology is business. Talwar was more specific on the mechanics of governance. He described a three-pillar approach: clarity of workflows, integrity of data, and a strong orchestration framework. On that third pillar, he said: As we scale AI agents across the enterprise, we treat them no differently than how we treat our human workforce. We treat them as a digital workforce that needs to be governed with the same rigor and policies as our human teams. The idea of AI agents as a digital workforce - subject to the same access controls, policies and accountability as human employees - is the underlying logic of everything ServiceNow is building around the AI Control Tower. Nvidia CEO Jensen Huang also joined McDermott on stage, and beyond the partnership details covered in the product piece, he also spoke about Nvidia itself being a ServiceNow customer. Huang said the company has cut employee intervention on support issues by two-thirds - two-thirds of queries that previously required a human are now resolved without one. And his advice on what enterprises should take from this was: Don't just think productivity and therefore cost reduction. Think productivity and therefore ambition elevation. He added: What used to take months, we think should take days. What we thought could take years, we now believe will get done this month. Underpinning all of this, McDermott outlined - without mentioning the dreaded 'SaaSpocalypse' phrase - how AI still needs deterministic workflows to execute on work effectively. Whilst the models are getting good at reasoning, he said that they still need the guardrails of a workflow. He explained: The world is converging around two critical technologies: AI that thinks, and workflow that acts. The first one gets all the attention - but execution is where enterprises actually win or lose. That is ultimately the pitch ServiceNow is making at Knowledge this week. The model providers are novel and are pulling focus at the moment, but ServiceNow believes that for serious enterprises, the models alone won't be able to execute work in a way that's trustworthy. McDermott also closed with a set of commercial guarantees for customers. He outlined a total satisfaction guarantee, described as a first for any enterprise software company; a go-live commitment in under 100 days; and AI Control Tower free for one year - which McDermott put a value of two million dollars on. Whilst this was framed as value for customers - which it certainly is - it also felt like a shot to the market and competitors. ServiceNow is putting its reputation on the line with firm commitments with its customers - what was left unsaid was the question of why others aren't doing the same. A clever move. I noted earlier today that Knowledge 2026 feels like a pivotal moment for ServiceNow's second reinvention - the shift from a workflow 'platform of platforms' to a work execution 'AI of AIs' company. The keynote this morning largely supported that, with McDermott speaking directly to enterprise AI challenges and offering a fuller answer to their problems. We will be following up on this pitch throughout the week with more thoughts from McDermott and the rest of the ServiceNow leadership team, as well as customers to provide the evidence - with the aim of understanding how buyers get from where they are to the vision outlined, as well as the challenges and opportunities ahead.

ServiceNow Knowledge 2026 - AI Control Tower expands, Autonomous Workforce reaches every function, and the acquisition strategy starts to add up

ServiceNow has announced an expansion of its AI platform at its annual user conference, Knowledge, in Las Vegas this week - with the AI Control Tower and its Autonomous Workforce seeing significant updates. The push with autonomous AI into every major enterprise function, together with new integrations from Microsoft and NVIDIA, as well as the launch of its Autonomous Security and Risk product, highlight how the vendor's recent acquisition strategy is starting to come together in the platform. At the center of the announcements is ServiceNow's latest platform release - Australia, which aims to take the AI Control Tower narrative beyond governance and into broader AI execution. Nenshad Bardoliwalla, Group VP of Product Management for AI at ServiceNow, explained in a pre-briefing how the announcements this week build on the foundations that ServiceNow laid last year: What we're hearing consistently from customers is that most organizations have more AI in production than they've inventoried or accounted for. And that anxiety around control, security and trust isn't going away - it's getting louder. The AI Control Tower has evolved significantly since we first introduced it at Knowledge in 2025. What we launched last year gave customers a governance layer, but what we're shipping this year goes significantly deeper, evolving from visibility and management into a full enterprise AI command centre. The point that ServiceNow will be aiming to get across this week is exactly that - that the AI Control Tower began as a governance layer, but has been reframed as a full enterprise AI command centre, now operating across five dimensions: Discover, Observe, Govern, Secure and Measure: On where most governance approaches currently fall short, Bardoliwalla said: AI agents are taking real actions, moving real money and affecting real people, but most of those agents are running without a system that governs them. That's not agentic business. That's agentic chaos. Interestingly, a new AI Gateway is also part of the Australia release, claiming real-time governance and observability controls for agentic workloads across any third-party AI system connecting to ServiceNow via MCP. It's always helpful to get a customer perspective amongst the product information - and Phil Priest, Head of Global Business Services at Rolls-Royce, joined ServiceNow Chief Customer Officer Chris Bedi to discuss AI deployment in practice across a business of 45,000 employees in 50 countries. Rolls-Royce launched Now Assist - branded internally as Merlin - in August 2025. The deployment has reached 12,000 employees and processes over 10,000 conversations a month. On the results to date, Priest said: It's delivered 5,000 hours of efficiency savings since we implemented it back in August 2025, with a 54 per cent deflection rate and a mean time to resolve down to two days. Priest noted that the 38,000 incidents deflected and resolved through predictive intelligence has translated into 300,000 saved shop floor hours - time returned to the engineers assembling Rolls-Royce engines. Bedi said: Translating that 54 per cent deflection in IT all the way through to 300,000 hours on the shop floor - that's where the rubber hits the road in a manufacturing operation. Priest also touched on the governance challenge. Using an accounts payable scenario, he walked through the complexity of deploying agents across a workflow that touches purchase order validation, invoice processing, payment execution and bank detail changes - areas where fraud risk is real and regulatory obligations significant. He said: Bank detail changes are an area where humans often sense that something doesn't look right... We have to think very carefully about building agents in a way that ensures fraud can't happen. He also offered a useful way of thinking about AI readiness more broadly, invoking a Bill Gates observation on automation: Automation applied to an efficient operation magnifies the efficiency, and automation applied to an inefficient operation magnifies the inefficiency... this applies to the use of AI - and the magnification occurs in seconds, almost straight away. Looking ahead, Rolls-Royce is planning to roll Now Assist across all services by October this year, with significant anticipated gains from onboarding and offboarding automation. Priest described the Moveworks acquisition as a particular source of anticipation for the business. He added: As you develop it further, we can't wait to see the additional benefits that can be delivered using this single-pane-of-glass approach across our entire enterprise and all of our content and services for our employees. I think it's a fantastic opportunity for us. Alongside the Control Tower expansion, ServiceNow is broadening its Autonomous Workforce into every major enterprise function. The L1 IT Service Desk AI Specialist - now generally available - is being joined by new AI specialists spanning IT operations, AIOps, site reliability engineering, asset lifecycle, portfolio planning, CRM, HR, finance, legal, procurement, workplace services, supplier management, health and safety and security operations. CRM AI specialists, covering sales qualification, quoting, order fulfilment, invoice disputes, service and renewals, are available now. IT AI specialists are expected in June; security and risk AI specialists enter preview in June with general availability targeted for September. All specialists run on shared platform infrastructure - the same operational intelligence via the CMDB and Context Engine, the same data connectivity via Workflow Data Fabric, and the same governance layer through AI Control Tower. Bardoliwalla described the distinction from task-based AI tools: These AI specialists execute complete workflows, carrying work from intent to outcome within a customer's guardrails. The security and risk announcements will draw increased attention this week, as the business line crossed $1 billion in annual contract value for ServiceNow last year, and the Autonomous Security & Risk launch represents the first full integration of both Armis and Veza into the platform. Armis - acquired for $7.75 billion, ServiceNow's largest acquisition to date - provides continuous, agentless asset intelligence across IT, OT, IoT, medical devices, cloud workloads and pre-compiled code, tracking nearly 7 billion connected assets in real time. Asset data flows enriched into the ServiceNow CMDB, turning a previously static inventory into a live picture of the attack surface. Veza's access graph sits alongside it, governing both human and non-human identities and enforcing least privilege at the point of action. John Aisien, SVP and General Manager for Security and Risk at ServiceNow, described the combined capability as an answer to the fragmentation problem CISOs have faced for years: There's detection in one stack, response in another stack, identity and permissions in a third, and asset visibility, if you have it, in yet another. The fragmentation - the seams between these different tools - are exactly what attackers exploit. Two new AI specialists handle vulnerability resolution and security operations end to end. ServiceNow's own security operations team is already running Autonomous Security & Risk, handling incidents seven times faster than prior workflows. On where the identity governance story is heading in the agentic era, Aisien said: In the same way that zero trust was a foundational security architecture for the cloud world, I expect zero permissions to become a foundational security architecture for the agentic world. And I think we're very well positioned to become one of the primary partners that makes that promise real. One other announcement that deserves a mention, and warrants its own follow-up, is ServiceNow's launch of Action Fabric. This opens the platform's full system of action to any AI agent in the enterprise via a generally available MCP Server - meaning agents built on Claude, Copilot, or a customer's own stack can now trigger governed ServiceNow workflows headlessly, without going through a traditional UI. ServiceNow has said that this is not about data access, it is governed execution - flows, playbooks, approvals and catalogue actions running through AI Control Tower, identity-verified and fully auditable. Anthropic is the first named design partner, connecting Claude Cowork directly into the ServiceNow system of action. The MCP Server is included in every Now Assist and AI Native SKU from today, with additional features expected in the second half of 2026. I'll follow up with a fuller analysis of what Action Fabric means in practice once I've had a chance to dig in further on the ground this week. Two further partnership expansions completed Tuesday's announcements. With Microsoft, ServiceNow is extending AI Control Tower governance across the Microsoft Agent 365 ecosystem - building on the November 2025 partnership that connected the Control Tower with Microsoft Foundry and Copilot Studio. ServiceNow AI specialists will be available in the Microsoft Agent 365 Marketplace, appearing in the org chart as digital employees with defined roles, permissions and accountability. The integration is available in preview; Marketplace availability is expected later this year. With NVIDIA, ServiceNow is introducing Project Arc - an enterprise autonomous desktop agent secured by the NVIDIA OpenShell sandboxed runtime and governed by AI Control Tower, available in early preview. AI Control Tower is also now included in the NVIDIA Enterprise AI Factory validated design, extending governance to large-scale model workloads at the infrastructure layer. The companies are also releasing NOWAI-Bench, an open benchmarking standard for AI agents comprising two frameworks - EnterpriseOps-Gym and EVA-Bench - as an open-source release. The pre-briefing and the product announcements give us a good flavor of what Knowledge 2026 has in store - but the real work starts this week on the ground, where we'll be talking to customers to understand impact, and pressing the vendor for a clearer sense of the direction of travel. There is plenty to get into. The product vision is ambitious. What ServiceNow is laying our here is a meaningful shift in language and intent - away from the automated workflows narrative it has built its business on, toward something it is now calling automated work. That is a subtle but significant distinction. Workflows are infrastructure. Work is what people actually do. Whether ServiceNow can make that claim credible across every enterprise function it has laid out this week is the question that will define the next chapter of the company's story. There are two other points that I'd like to get an understanding on this week. The first is the blueprint for getting from here to there. The Autonomous Workforce vision - AI specialists resolving cases, containing threats, fulfilling orders and processing invoices end to end, across IT, HR, finance, legal, procurement and security - is no doubt compelling. But enterprises are not monolithic. They are collections of competing organisational silos with different agendas, different budgets and different appetites for change. How ServiceNow navigates that reality in practice, what the implementation path actually looks like, and where the blockers are - those are the conversations I am hoping to have with customers on the ground. The second is competitive positioning. ServiceNow is not making this argument alone. Microsoft, Salesforce and Google Cloud are all fighting for versions of the same thesis - that their platform should be the one that governs, orchestrates and ultimately executes enterprise AI at scale. My instinct, reading between the lines of what Bardoliwalla and Aisien were saying, is that ServiceNow's bet is a specific one: that if it can win on governance and trust first, the rest of the platform argument follows. I want to get further proof points from the event this week that that logic is sound for enterprise buyers.