has an execution problem
AI hasn't delivered a significant return on investment (ROI) for most, and it’s not a budget problem. AI spending surged by 110% this year. It’s not a conviction problem either. Agentic AI is the most significant factor shaping technology strategy right now, and most executives believe autonomous operations are no longer optional but rather a competitive necessity.
The problem is execution. According to our panel of 100 executives, all deeply involved in AI and enterprise technology strategy, there are five specific barriers holding organizations back.
Most executives we surveyed need to prove AI value fast. They’re chasing visible, quick wins and near-term financial impact to gain stakeholder confidence.
This expectation is shaping how AI strategies get built. Cost reduction and efficiency are the primary drivers of AI strategy for many, well ahead of market opportunity and business resilience.
Despite those ambitions, few report significant ROI, and escalating AI costs can make cost efficiency a moving target. Meanwhile, few formal commitments over the next 12 months focus on equally important outcomes such as revenue growth and reducing risk exposure. The result is a narrow investment posture that favors short-term gains over transformation.
From insight to action
Move beyond efficiency with orchestration. Efficiency is a legitimate starting point, but it shouldn't be the destination. When organizations focus on automating pockets of siloed processes, individual tasks get faster, but the gains are incremental. Moving beyond that requires organizations to embed AI across departments and connect it with a single layer that coordinates decisions and outcomes across the enterprise.
That shift from automation to orchestration unlocks outcomes that cost efficiency alone won’t yield. Organizations should set explicit commitments around revenue growth and risk reduction and align AI investments accordingly. The compounding returns come from breadth, not from going deeper on a single use case.
The executives on our panel know where they want to go. Most agree that becoming a more autonomous, self-optimizing enterprise is a competitive necessity, not merely an operational improvement. But few have moved beyond select agentic AI use cases or pilots. Fully autonomous operations remain rare. That’s a long way from their goal of having a coherent, AI-powered operating layer that can act across the enterprise.
Data readiness and system fragmentation are the two barriers that keep surfacing, and they’re deeply connected. The underlying constraint is the same in both cases: AI needs clean, unified data and connected workflows to orchestrate decisions and outcomes across departments and systems at scale—the panel's top priorities for deploying and scaling AI effectively.
From insight to action
Build the AI foundation. Fragmented systems and siloed data don't just slow deployment. They limit what AI can do once it's deployed. Building a clean, governed, and connected data foundation needs to be treated as a strategic prerequisite for making AI effective at scale.
Organizations should consolidate that foundation onto integrated platforms that connect AI-enabled workflows across departments and systems. That shift is what makes true autonomy possible, giving AI the right data, context, and infrastructure to act at the moment it matters.
Governing AI is one of the top challenges faced by IT teams, according to our panel, beating out other demands such as budget pressures and talent retention.
The biggest governance gaps are data privacy, security exposure, and third-party AI vendor risk. But as organizations push harder to scale AI, a meaningful share says current frameworks are slowing deployment.
As AI takes on more decision-making, the absence of clear governance frameworks amplifies risk. The ServiceNow Enterprise AI Maturity Index 2026 points to why: Only 26% of organizations have established systems to manage governance and compliance.
Governance issues extend to questions of responsibility. Most say the chief information officer or IT leadership holds primary responsibility. A smaller group has distributed it to individual business units. A few have not yet established ownership at all.
As agentic AI moves into operations, finance, security, and customer-facing functions, that accountability fragmentation may become a structural risk.
From insight to action
Accelerate governance. Close the chaos gap. Treat governance as a competitive advantage, not a compliance burden. That means moving beyond one-time policy documents toward automated processes: continuous regulatory scanning, ongoing risk assessment, and enterprisewide compliance tracking.
Organizations should establish a cross-functional governance structure with clear authority to set standards and intervene when risks emerge, rather than having distributed accountability across business units without a coordinating layer. As AI moves into every business function, that structure becomes essential. Organizations should also define exactly where human judgment begins, as autonomous doesn't mean unattended.
Across all business outcomes, the panelists say they’re least confident about proactively preventing security breaches and managing cyber risk. It's also one of the most active areas for autonomous use cases, second only to service operations.
The threat environment explains why this makes sense. AI-powered cyberattacks, third-party AI vendor risks, and data exposure from AI agents accessing sensitive systems are top security concerns.
The ServiceNow 2026 Risk and Security Outlook found that attack capabilities are outpacing organizational defenses, with most executives reporting only low to moderate confidence in their ability to handle AI-powered threats across every major attack vector. The volume and sophistication of threats have made autonomous security capabilities a practical necessity.
From insight to action
AI threats need autonomous defenses. The threat environment has outpaced what human-led security operations can realistically manage at scale. The volume, speed, and sophistication of AI-powered attacks have made autonomous defense capabilities a practical necessity, not a future state to plan toward.
Organizations should close the gap between detecting a risk and acting on it through automated workflows that route ownership, trigger remediation, and document every action without manual handoffs. Third-party AI vendor risk demands the same discipline: continuous monitoring, not quarterly reviews. And as AI agents access sensitive systems with permissions that, in many cases, no one has fully mapped, that governance gap needs to close before it becomes a breach.
IT is no longer focused on just operational support. It’s evolving toward strategic value delivery. That shift is generating real pressure, with IT caught between competing demands. Proving ROI remains IT’s biggest challenge, according to the panel, and their biggest challenge is balancing innovation and AI experimentation while keeping existing operations stable.
This tension is shaping how IT leaders plan. The 12-month vision for many is a more proactive, hybrid autonomous model, where AI handles high-volume, lower-risk activities while humans own critical decisions requiring judgment. Service operations and security operations are already the panel’s most active area for autonomous deployment, reflecting where volume and complexity have outpaced what human-led operations can realistically support.
From insight to action
Free IT teams to innovate. For many IT leaders, the biggest obstacle to innovation isn't capability; it's capacity. Organizations should prioritize automating high-volume, routine operations not just as an efficiency play, but to buy back the time IT needs to work strategically. Without that, innovation stays on the roadmap indefinitely.
As IT takes on greater strategic responsibility, organizations should redefine how their value is measured. Success should shift from operational metrics toward the business outcomes IT is responsible for driving.
The gaps these 100 leaders are signaling are confirmed at scale in the ServiceNow Enterprise AI Maturity Index 2026 across 4,500 executives globally. Most organizations are automating yesterday's work rather than reimagining tomorrow's. Efficiency is the baseline outcome everyone is chasing, and transformation is the ultimate goal they all aspire to.
The organizations that close the gap between ambition and execution are the ones that build their strategy and infrastructure accordingly.
The Exec Signals panel comprises 100 technology and business executives from large global companies, all with decision-making or significant influence over AI and enterprise technology strategy. Insights were gathered through a structured survey, surfacing directional intelligence on how the leaders are navigating AI. Research was conducted independently by global research firm Phronesis Partners.
