ServiceNow’s India Bet: Can It Put AI To Work Inside Enterprises?

Enterprise software major ServiceNow is betting that its next leg of growth in India will come not just from digitisation of IT functions, but from enterprises putting AI to work across business processes, even as the rush to adopt the technology creates a parallel challenge around governance, security and identity.

At its World Forum in India on Tuesday, October 6, ServiceNow announced AI Workflow Factory and Autonomous Engineer, two products aimed at helping enterprises identify where AI can improve work and then build, run and extend AI-powered workflows across the organisation.

The launches come at a time when Indian enterprises are moving beyond AI experimentation, with CIOs increasingly looking at how to deploy the technology across customer experience, employee operations, software development and business processes.

For ServiceNow, that creates an opportunity to move beyond its traditional identity as an IT service management company and position itself as the layer through which enterprises orchestrate AI across their existing technology stack.

But the company is also betting that as AI becomes more deeply embedded into enterprises, security and governance will become inseparable from the adoption story.

ServiceNow’s India AI Push Moves Beyond Pilots

ServiceNow’s India strategy rests on three interconnected opportunities: domestic enterprises, global capability centres (GCCs) and its large technology partner ecosystem, ServiceNow India tech and business centre head Sumeet Mathur said in an interview with Inc42.

The company has been expanding beyond its traditional strength in IT service management, with ambitions to become an “enterprise operating system” spanning IT, HR, customer service, finance, supply chain, legal and workplace operations.

AI is now becoming the connective tissue across these functions. The company is seeing a growing willingness among Indian CIOs and CISOs to explore disruptive AI use cases rather than simply run small pilots.

“The willingness is very high, and once you start to open your mind that I want to use this piece of technology in 10 different ways, then of course the problems are very visible,” Mathur said.

That appetite, however, is running ahead of the governance structures needed to deploy AI safely. A ServiceNow AI maturity assessment cited by the company found that only 22% of Indian enterprises have AI testing, auditing and risk-assessment processes in place.

Amit Zavery, president, COO and chief product officer at ServiceNow, said on the sidelines of the event that governance cannot become an excuse for delaying AI adoption but needs to be built into the process early.

“It just slows you down enough to run faster later. If you’re going to run faster without any kind of brakes around it, it will crash,” Zavery said.

The distinction is becoming increasingly important as enterprises move from individual AI experiments to multiple deployments across an organisation. The question is no longer simply whether an AI application works, but whether it can be governed, monitored and scaled without creating new operational or security risks.

From IT Workflows To An Enterprise Operating System

ServiceNow’s pitch is that enterprises do not need another standalone AI application. They need a layer that can connect AI models, existing applications, data and workflows.

AI Workflow Factory is designed to help enterprises identify opportunities where AI can improve workflows and then build and operate those workflows. Autonomous Engineer, meanwhile, brings more autonomous capabilities into software development, including planning, building and testing.

The proposition addresses one of the biggest problems emerging as companies scale AI: fragmentation. Enterprises building AI applications often have to stitch together different models, data sources, applications and workflows. They also have to manage changes when models are updated or prompts and underlying systems change.

AI Workflow Factory brings more of that lifecycle onto a common platform, including building, deployment, governance and ongoing management.

The broader ambition is to make ServiceNow the operating layer through which enterprises can introduce AI into existing business processes without having to rip out their existing technology infrastructure. This is also where ServiceNow believes it can differentiate itself from hyperscalers and model providers.

Microsoft, Google and AWS can provide infrastructure and AI models, while companies such as OpenAI and Anthropic provide frontier models. ServiceNow, however, wants to sit one layer above these technologies, connecting them to the enterprise systems where work actually happens.

AI Agents Create A New Identity Problem

That strategy becomes more complicated as enterprises move from generative AI assistants to autonomous AI agents. Today, many enterprise agents still operate on behalf of a human. But as agents begin interacting directly with email, CRM systems, databases and other enterprise applications, the identity and access problem becomes significantly more complex.

An AI agent could initiate another agent, access sensitive data and execute an action without a human manually intervening at every step.

“The hardest problem I believe for agentic AI to become a reality is identity,” said Tarun Thakur, cofounder and CEO of identity security startup Veza, which ServiceNow acquired earlier this year. Veza was founded around the problem of understanding not just who an employee is, but what that identity can actually access across an enterprise’s applications and systems.

Its Access Graph maps identities, permissions and relationships across enterprise systems, giving security teams visibility into effective permissions rather than simply relying on roles and groups.

ServiceNow historically had extensive visibility into enterprise assets through its configuration management database (CMDB), but lacked the same depth of access context. The Veza acquisition brings those two pieces closer together — what an enterprise has and who or what can access it.

That becomes particularly important as the definition of an “identity” changes. “Agent identity is ephemeral — it comes and goes,” Thakur said, pointing to a future where enterprises could have large numbers of temporary machine identities being created and removed as AI agents perform specific tasks.

ServiceNow has also integrated Veza into its broader security portfolio, while using its AI Control Tower to help enterprises discover and manage AI agents across their technology environment.

The company believes AI security could eventually become a significant new cybersecurity category as enterprises move towards more autonomous systems.

The India AI Opportunity

The AI opportunity also gives ServiceNow a broader opening in India than its traditional IT-service customer base. India is home to more than 2,100 GCCs employing around 2.35 Mn people, according to Mathur. These centres are increasingly responsible for technology, product development, cybersecurity and global technology decisions for their parent companies.

For ServiceNow, that means the India opportunity can extend beyond selling to Indian businesses. GCCs can also influence technology decisions made for global organisations.

The company’s partner ecosystem further strengthens that position. Mathur said around 30,000 engineers are working on ServiceNow in India across major system integrators including Accenture, Deloitte, KPMG and EY.

BFSI remains among the company’s strongest verticals in India, while healthcare and life sciences, consumer and retail and the public sector are also important markets. Interestingly, in July, ServiceNow’s venture arm invested $40 Mn in Delhi NCR-headquartered BUSINESSNEXT, which sells software to banks, insurers and other financial institutions.

Betting On The Orchestration Layer

The larger strategic question is whether ServiceNow can establish itself as the control and orchestration layer as enterprises adopt AI from multiple vendors. The company is not trying to own the underlying AI models. Instead, it is building an open platform that can work with different frontier, open-weight and domain-specific models.

Zavery said customers ultimately care about the outcome rather than which model powers a particular task. ServiceNow therefore tests different models and can route workloads depending on the task, performance and economics.

That model-agnostic approach could become increasingly important as enterprises seek to avoid being locked into a single AI provider. ServiceNow is also competing in a market where Microsoft, Google, AWS and Salesforce increasingly position themselves as orchestration or control layers for enterprise AI.

According to Mathur, ServiceNow’s advantage comes from its existing position across enterprise workflows. “ServiceNow was always a platform company,” he said, arguing that its role is to coordinate different systems and business processes rather than own every underlying technology.

That positioning is increasingly extending into cybersecurity as well. ServiceNow’s security business was already generating more than $1 Bn in annual contract value globally before its recent acquisitions, according to Mathur, and the company is now seeking to build a much larger cybersecurity franchise around identity, vulnerabilities, incident response, OT/IoT and AI security.

The Veza acquisition therefore fits into the same broader thesis: as more AI agents enter the enterprise, ServiceNow wants visibility into not just the applications and workflows they interact with, but the identities and permissions behind them.

For Indian enterprises, the next phase of AI adoption could therefore shift the question from what AI can do to what AI can do inside their organisation, especially at scale. That is the gap ServiceNow is betting its India strategy can fill.

The post ServiceNow’s India Bet: Can It Put AI To Work Inside Enterprises? appeared first on Inc42 Media.