ServiceNow's CIO Kellie Romack on the Evolving Role of Humans in AI-Powered Workflows

ServiceNow's AI-Powered IT Support Transformation

ServiceNow, a leading enterprise software firm, has taken a unique approach to automating its internal IT support operations using artificial intelligence (AI). Instead of cutting most jobs, the company opted for reassignment, moving employees into newer roles tied to AI operations, analytics, and security. According to Kellie Romack, ServiceNow's chief digital information officer, nearly 85% of the IT help desk workforce was redeployed after AI agents took over a significant portion of their work.

Romack, who leads internal technology operations and AI transformation at ServiceNow, emphasized that the company conducted thorough internal assessments to understand employee skill sets and interests before moving them into adjacent functions. Some employees shifted into AI operations and security-related work, while others moved into executive briefing centers and data analysis roles. The remaining help desk staff were retrained to work alongside AI systems rather than compete with them.

$380 Million in Productivity Gains

Read also: SpaceX Seeks Record $75 Billion IPO, Potentially Positioning Elon Musk as the World's First Trillionaire

ServiceNow has been at the forefront of deploying AI internally, with the company previously disclosing productivity gains worth roughly half a billion dollars from its AI initiatives. Romack stated that around $380 million of that value came from ServiceNow's own software stack. The company's IT support teams now use dashboards and monitoring systems to supervise AI agents, monitor workflows, and intervene when systems fail or require escalation.

The Importance of Human Oversight

Romack emphasized the critical role of human oversight in AI deployments. "Humans create AI. It's our job to create governance for AI," she said. "We are accountable and responsible for it." She cautioned that companies deploying AI systems cannot afford to treat them as opaque black boxes operating without checks and balances. "We talk a lot about moving AI from a black box to a glass box," she said. "People fear what they do not understand."

Best Practices for AI Deployments

Read also: Treasury Yields Experience Largest Increase in Two Weeks Following Release of Labor Market Data

Romack also shared guidance on best practices for AI deployments. She warned against simply layering generative AI tools on top of outdated workflows without redesigning the underlying process itself. "Do not put AI on top of an old process," she said. "Reinvent the work and the way things are done." She emphasized the need to combine business context, security systems, and governance frameworks within a single architecture, especially as companies begin adopting autonomous AI agents.

Note: The comparison of ServiceNow's productivity gains from AI initiatives is presented in a Markdown table.