ServiceNow Charts a Course for Autonomous AI, Even as Governance Doubts Linger

ServiceNow pivots to autonomous AI agents, reports 22% revenue growth and raised guidance, but faces governance challenges as trust in unsupervised AI remains low.

For a company whose stock has been drifting sideways for weeks, ServiceNow’s latest strategy shift is anything but timid. At the Knowledge 2026 conference, Chief Product Officer Amit Zavery declared that advisory artificial intelligence has “reached its end” — and that the future belongs to AI that acts, not just suggests. The claim is bold enough to unsettle any enterprise customer worried about handing the keys to a machine.

Yet ServiceNow is leaning in. It has expanded its “Autonomous Workforce” with AI agents that independently handle IT support, HR inquiries, customer service tickets and security threat responses. In an internal trial, the company’s own AI specialist resolved IT cases 99% faster than human staff. The pitch is less about efficiency gains and more about a fundamental redefinition of how work gets done: these agents sit alongside employees rather than inside a chat widget.

Clients are already seeing results. The city of Raleigh cut its IT support costs by 66%, while Honeywell sped up compliance verification by 75%. Nvidia CEO Jensen Huang lent his credibility to the narrative during a keynote, calling ServiceNow the “operating system for enterprise AI agents”. The company’s ambition is to become the control tower that monitors every autonomous action across an organisation — a positioning it brands as the “AI Control Tower for Business Reinvention”.

That message resonated enough to attract a new ally. ServiceNow announced a partnership with Hitachi Digital Services to use AI for monitoring critical IT infrastructure, aiming to reduce downtime. The alliance fits squarely into the platform’s broader strategy of embedding itself as the central nervous system for enterprise workflows — a role it already plays for 85% of Fortune 500 companies.

But autonomy raises uncomfortable questions, and ServiceNow is not dodging them. The company’s chief customer officer acknowledged bluntly that decision rights must be clarified before scaling: who approves what an agent can do, how far its autonomy extends, and crucially, who bears liability when something goes wrong. A survey from the AI Builder Summit 2025 found that only 44% of enterprise AI leaders trust agents to act without human oversight. This governance challenge could become the defining battleground of the coming decade in enterprise software.

Financially, the story remains compelling. First-quarter 2026 subscription revenue rose 22% to $3.671 billion, matching total revenue growth. Management raised its full-year guidance from roughly $15.55 billion to $15.735 billion. The AI product portfolio is on track to reach nearly $1 billion in annual contract value by year-end. Yet the share price — recently around €95 after dipping to €93.68 on the day of the Hitachi news — trades well below the average analyst target of €123.38, implying roughly 30% upside. The RSI sits near 55-57, firmly neutral, while 30-day volatility of over 67% suggests the market is waiting for a decisive catalyst.

That catalyst may come on July 22, when ServiceNow reports second-quarter results after the US market close. Investors will be listening closely to the accompanying webcast for evidence that the AI initiatives are translating into sustained revenue acceleration. For now, the company has momentum, a fresh partnership and a clear — if high-stakes — vision. The missing piece is the confidence that autonomous agents can be trusted at scale.