Ninety percent of companies anticipate a shift toward autonomous IT operations within the next two years, envisioning agentic AI systems that independently plan and execute multi-step tasks. However, a significant majority, 77 percent, express hesitation in allowing AI to make operational decisions without human approval. This suggests that most organizations currently aim for a model where a human remains involved in the approval process, rather than fully autonomous systems.
Despite these aspirations, many companies are not yet prepared for such a transition. Only 41 percent report their IT environments are ready for autonomous operations, and a mere 19 percent of their IT operations are currently automated. This indicates a disconnect between strategic planning and current operational capabilities.
A key point of divergence exists between business and IT leaders and the technical specialists who implement these systems. While leaders are more optimistic about the adoption of autonomous IT, technical specialists are warier, particularly regarding the foundational groundwork. Specialists also tend to rate their organization's data readiness lower than leaders, suggesting a more realistic view of the challenges involved.
A critical obstacle to AI-driven IT is the lack of comprehensive visibility across the entire IT landscape. A vast majority of respondents, 96 percent, emphasize the importance of unified visibility across networks, applications, endpoints, and cloud environments for effective AI diagnostics. Yet, only 17 percent currently possess such unified visibility, with data often fragmented across silos. This issue is exacerbated by tool sprawl, which most companies acknowledge increases the need for system integration, despite also recognizing that consolidation would reduce operational friction.
Security and compliance concerns are the primary obstacles cited, with leaders weighing these issues most heavily. Technical specialists, on the other hand, more frequently highlight fragmented tools as a significant challenge. Other impediments include skills gaps and resistance to change within organizations.
Furthermore, many companies report insufficient oversight of AI agent performance and reliability, with 51 percent indicating a lack of adequate monitoring. There is also a general lack of understanding regarding employee AI usage and associated costs. Technical specialists express less confidence than leaders in the effectiveness of their AI governance, providing a concrete reason for maintaining human approval in operational decisions.
Looking ahead, nearly all respondents expect AI to transform the functions of frontline IT and service desk teams within two years. Anticipated changes include a reduction in overall service desk activity, a shift toward handling more complex issues, and a transition from reactive problem-solving to proactive prevention. Very few expect the service desk to remain largely unchanged. Already, a third of companies report their IT teams are highly effective at using AI to identify and resolve issues before employees notice them.
