Artificial intelligence has quickly moved from experimentation to everyday operations across enterprise IT environments. From automating ticket resolution and triaging incidents to identifying system issues before they affect users, AI is increasingly being positioned as the technology that will transform IT service management (ITSM). Yet, according to new research from SolarWinds, the reality of AI adoption is far more complex than many organizations anticipated.
The company’s 2026 State of ITSM Report, based on a survey of more than 800 IT professionals worldwide, reveals a striking paradox: while AI is largely delivering on its promises of efficiency and return on investment, it is not necessarily making life easier for IT teams. Instead, many organizations are discovering that AI is creating entirely new categories of work, introducing operational complexity, and generating unexpected costs that were rarely accounted for in early business cases.
The findings suggest that the conversation around AI in ITSM is evolving. The challenge is no longer convincing organizations to adopt AI. The real challenge lies in ensuring that AI delivers meaningful outcomes without overwhelming the teams responsible for managing it.
The Good News: AI is Delivering Results
At first glance, the report paints an encouraging picture of AI adoption in IT service management.
An overwhelming 84% of surveyed professionals said AI has either met or exceeded their return-on-investment expectations, a clear indication that organizations are seeing measurable value from their investments. AI-powered tools are helping service desks streamline routine operations and automate repetitive tasks that have traditionally consumed a significant amount of IT staff time.
Respondents reported that AI saves an average of:
- 3.2 hours per week on issue detection and flagging
- 3.0 hours per week on managing end-user requests
- 2.9 hours per week on ticket triage and prioritization
These efficiency gains may appear modest on an individual basis, but when multiplied across large enterprise IT teams, they represent substantial productivity improvements. Faster issue identification, automated workflows, and intelligent routing of incidents allow organizations to respond quicker and improve service levels while reducing manual intervention.
For business leaders who have spent the past two years pouring investments into AI initiatives, these numbers validate the belief that intelligent automation can drive tangible improvements in operational performance.
The Hidden Cost of Efficiency
However, beneath those positive results lies a more complicated reality.
While AI is reducing time spent on traditional ITSM activities, it is simultaneously creating entirely new responsibilities. The report found that 52% of IT professionals say their workload has actually increased since implementing AI.
Rather than replacing work, AI is often reshaping it.
Nearly half of respondents indicated they now spend considerable time:
- Managing AI tools and integrations (48%)
- Reviewing and validating AI-generated outputs (47%)
- Training and fine-tuning AI models (37%)
These tasks rarely existed at scale before AI entered the workplace. Today, they have become a routine part of IT operations.
In many organizations, teams are discovering that AI systems cannot simply be deployed and forgotten. Models require monitoring, continuous training, governance oversight, and ongoing optimization to ensure outputs remain accurate, relevant, and compliant with business requirements.
This growing maintenance burden has created what many industry observers describe as the “AI productivity paradox” where technology delivers efficiency gains while simultaneously increasing overall operational complexity.
The SolarWinds study suggests that many enterprises are experiencing exactly that phenomenon.
“We’re at an inflection point in IT service management. AI adoption is no longer the hard part. The hard part is building the organizational discipline to make AI actually deliver. The teams that get this right aren’t just running a faster service desk; they’re running a fundamentally different operation.”
Brad McGinity, GM of ITSM, SolarWinds
Why AI Costs More Than Expected
Cost overruns represent another major theme emerging from the report.
Although organizations are generally satisfied with AI’s performance, very few anticipated the true cost of adoption.
Only 7% of respondents said the cost of implementing AI matched their original expectations.
Instead, organizations reported unexpected spending in three critical areas:
- Staff training and skills development (48%)
- Data quality improvement and remediation (47%)
- Ongoing AI tuning and maintenance (45%)
These expenses can significantly alter the economics of AI projects.
Many AI business cases focus heavily on software licensing and implementation costs while underestimating the resources required to prepare data, train employees, and establish governance frameworks. The result is that organizations often achieve technological success but encounter financial realities they did not plan for.
The survey also found that 83% of organizations spend at least three hours each week maintaining AI systems, highlighting the fact that AI introduces a recurring operational commitment rather than a one-time deployment effort.
As AI becomes embedded within core business processes, maintenance and oversight are increasingly emerging as permanent line items in enterprise IT budgets.
AI Remains Largely Reactive
Perhaps the most intriguing finding from the report concerns how organizations are actually using AI.
Despite years of discussion around predictive analytics, autonomous operations, and self-healing environments, most ITSM teams still rely on AI primarily for reactive activities.
When respondents were asked where AI has delivered the greatest impact during the incident lifecycle, the leading responses were:
- Identifying issues before users are affected (31%)
- Prioritizing and routing incidents (23%)
While these capabilities certainly improve efficiency, they are still focused on responding to problems that have already emerged.
Only 19% of respondents identified preventing incidents before they occur as AI’s greatest contribution.
This suggests that many organizations remain at the early stages of AI maturity.
The technology itself may be capable of far more advanced capabilities, but the supporting infrastructure, data quality, governance frameworks, and organizational processes required to unlock proactive operations often remain underdeveloped.
As a result, many enterprises are using AI to become better firefighters rather than eliminating fires altogether.
The Maturity Gap
The report points to a widening gap between AI adoption and AI maturity.
Over the past few years, organizations have focused heavily on deploying AI tools. Today, that phase appears largely complete. The challenge now is creating the operational environment necessary to maximize their value.
Successful AI adoption requires more than algorithms and automation platforms. It depends on high-quality data, integrated workflows, clear governance structures, employee training programs, and carefully defined performance metrics.
Without these foundational elements, AI risks becoming another technology layer that adds complexity instead of reducing it.
SolarWinds argues that many organizations are currently operating in this transition phase, where AI has been introduced into existing workflows but has not yet fundamentally transformed how service management operates.
The result is a period of adjustment during which organizations must simultaneously manage traditional IT responsibilities and new AI-related obligations.
Investment Continues to Accelerate
Despite these challenges, organizations show no signs of slowing their AI investments.
The survey found that 85% of organizations increased their AI budgets for ITSM compared to the previous year, while 36% reported significant increases.
This continued investment underscores growing confidence in AI’s long-term potential, even as short-term challenges persist.
Particularly noteworthy is the growing focus on agentic AI workflows, which emerged as the area expected to receive the highest future investment.
Unlike traditional automation tools that execute predefined tasks, agentic AI systems can make decisions, coordinate actions across multiple systems, and proactively address issues with minimal human intervention.
Many technology leaders view agentic AI as the next stage in the evolution of IT operations, enabling a shift from reactive service management toward predictive and autonomous environments.
Building a Better AI Strategy
The report concludes that organizations must become more deliberate in how they approach AI adoption.
Rather than deploying AI broadly and hoping productivity improves, enterprises should focus on targeted, measurable implementations.
SolarWinds recommends beginning with high-volume, clearly defined processes such as ticket triage, incident documentation, and issue detection. These areas provide clear success metrics and rapid feedback loops, making it easier to demonstrate value.
The company also emphasizes the importance of consolidating AI within existing service workflows instead of introducing disconnected tools that create additional maintenance burdens.
Data quality remains another critical factor. Poor-quality data continues to be one of the leading reasons AI projects underperform. Organizations that view data management as a central element of AI strategy are more likely to achieve sustainable outcomes.
Perhaps most importantly, the report encourages organizations to measure outcomes rather than activity. Teams that focus on business results, user experience improvements, and operational performance are better positioned to realize meaningful returns than those tracking AI usage metrics alone.
The Road Ahead
The SolarWinds 2026 State of ITSM Report paints a picture of an IT industry entering a new phase of AI adoption. The technology is delivering measurable benefits, validating years of investment and experimentation. However, organizations are also learning that successful AI implementation requires far more than deploying advanced tools.
The next chapter of AI in IT service management will not be defined by adoption rates. It will be defined by maturity, governance, data quality, and operational discipline.
For IT leaders, the message is clear: AI can transform service management, but only when organizations invest as heavily in strategy, processes, and people as they do in the technology itself. Until then, many IT teams may continue to experience a surprising reality where AI makes them more productive, yet busier than ever before.
