ServiceNow AI Workflow Factory Targets India's Agent Gap

What ServiceNow announced

ServiceNow used its World Forum in Mumbai on October 6 to unveil two products aimed at moving enterprise AI from experiments into everyday operations: AI Workflow Factory and Autonomous Engineer.23 The company says the tools help organisations find where AI can improve work, then build, run and extend AI-driven workflows.3

The two products are at different stages. AI Workflow Factory is generally available worldwide now. Autonomous Engineer, a coding agent designed to work unattended, is in early access on request.23 ServiceNow did not announce pricing.2

A bundle arranged as a loop

AI Workflow Factory is mostly a new arrangement of existing ServiceNow products, with some new pieces added.2 The steps work as a cycle:

• Process Mining identifies processes worth changing, tied to KPIs that business units already track.
• Autonomous Engineer and Build Agent build the change.
• App Engine runs it at scale.
• AI Control Tower governs the AI at every step.
• Action Fabric connects third-party agents and tools outside the ServiceNow platform.2

Coverage aimed at investors stresses that this links process mining with autonomous engineering. The point is for workflows to keep improving with less manual effort once they are digitised.5 That analysis also names the commercial logic plainly. Continuous, agent-driven refinement makes workflow subscriptions stickier and more valuable.5 Reports focused on the Indian market emphasise a different promise: closing execution gaps caused by legacy infrastructure and letting businesses scale operations in days.1

Autonomy, with a checkpoint

The name "Autonomous Engineer" claims more independence than the product actually has. According to one analysis, the agent plans, builds and tests implementation work by itself, but developers keep control of critical decisions before anything reaches production.4 That analysis argues that human review is the normal pattern wherever automated outputs lead to real-world action. Systems that make consequential calls without a human checkpoint are the exception.4

This design matters for buyers. Enterprises that hesitate to let an agent ship code will find the review gate reassuring. It also means the productivity gains depend on how quickly human approvers can keep up.

The India numbers behind the pitch

ServiceNow put India at the centre of the launch, and that was deliberate. Management described the country as a key hub for building and deploying these capabilities for global customers.5 Indian partners are reportedly already adopting AI Workflow Factory and using Autonomous Engineer for unattended coding.1 The company also points to research suggesting Indian enterprises are putting AI into operation faster than the global average.1

The same research shows the size of the gap ServiceNow is trying to close. A survey of 4,500 senior leaders found that 54% of Indian organisations are deploying AI agents, but only 11% have moved to autonomous workflows.3 It also projects that AI will take up more than 21.3% of the average Indian IT budget by 2027, up from 16.6% today.3

The sources frame these figures differently. Some coverage leads with India's speed relative to the rest of the world.1 Other coverage leads with the drop-off between deploying agents and actually running autonomous processes.3 Both readings can be true. India may be ahead of the global average and still far from its own ambitions.

Why it matters

The 54-to-11 gap is a useful reality check for enterprise AI generally. Deploying an agent is easy to report as progress. Rebuilding a process so the agent runs it end to end, under governance, is much harder. ServiceNow's loop is designed for that harder step. It pairs discovery, through process mining, with control, through AI Control Tower, so that pilots can turn into production workflows.

Budget data shows where the money is going. If AI's share of Indian IT spending grows by about a third over the next couple of years, vendors that can show measurable workflow outcomes will gain most.3 Tying Process Mining to existing business KPIs is an attempt to give customers ROI they can measure.2

The reading

This launch is less a technical breakthrough than a packaging and positioning move. It bundles familiar components into a story about execution and anchors that story in the market with the clearest gap between enthusiasm and results. Several things remain unknown: pricing, how far Autonomous Engineer gets beyond early access, and whether customers can show that agents are moving organisations from the 54% group into the 11% group.23

If ServiceNow can document that shift with real customer results, the India-first strategy will look smart. Until then, the survey figures say as much about how hard autonomous workflows are as they do about demand for them.