The state of AI in service management: driving efficiency across IT and beyond

Artificial intelligence (AI) is rapidly reshaping how organizations manage services across various business functions such as IT, HR, and R&D. This article highlights insights from a recent ITIL webinar featuring Aditya Mani, Senior Product Manager for ITSM at Atlassian; Kate Clavet-D’Amelio, Senior Technical Product Marketing Manager for ITSM at Atlassian; and Dmitry Isaychenko, Portfolio Director at PeopleCert.

The discussion centred on key findings from the State of AI in Service Management report, developed by Atlassian, an ITIL-accredited Tool Vendor. The report is based on survey responses from IT, HR, customer service, and product teams in companies with 500 or more employees. It explores three main areas: AI adoption, challenges and barriers to its implementation, and future investment priorities.

This article highlights key trends identified in the report, real-world applications of AI, levels of maturity, challenges faced, and the role of ITIL in supporting responsible AI adoption.

The adoption of AI in service management has accelerated significantly in recent years. Last year, about 16 per cent of organizations reported that they were not using AI. However, the latest survey revealed that only two per cent remain entirely disengaged. Most teams are now in the piloting phase, experimenting with AI to optimize their workflows and many are already in the optimization phase, refining what they have implemented.

Organizations are recognizing that AI is particularly effective for repetitive tasks, allowing employees to focus on more creative and higher-value work. In IT operations, AI can automate tasks such as infrastructure monitoring and scaling, including spinning up new servers or handling clusters autonomously. Meanwhile, HR teams are leveraging AI-powered onboarding agents to assist new hires in navigating resources and knowledge bases, reducing manual operations by 60 per cent and saving hiring managers an average of four hours per employee. In R&D, AI helps accelerate experiment planning, automate literature reviews and identify patterns in complex datasets, freeing research teams to focus on hypothesis-driven work. Across the board, AI is seen as a tool to complement, rather than replace, human effort.

Driving efficiency and knowledge management

The benefits of AI extend far beyond just automating tasks. One significant area where AI adds value is knowledge management. In large enterprises, AI helps uncover hidden patterns, generate accurate outputs from extensive data sets, and streamline the curation of information. IT operations teams use AI to detect issues, reduce Mean Time to Resolution (MTTR), and prevent incidents. Meanwhile, customer service teams benefit from virtual agents that can deflect up to 30 per cent of routine requests.

Companies such as U.S.-based company Wex have reported a 25 per cent increase in automation, a 17 per cent boost in administrative productivity, a 20 per cent improvement in MTTR, and a 60 per cent reduction in ticket escalations. AI also leads to substantial time savings: IT operations teams report an average of 55 minutes saved per incident and a 35 per cent faster approval rate for change requests. Results such as these highlight the tangible impact of AI on productivity, efficiency, and workflow optimization.

Addressing challenges and barriers

Despite the growing adoption of AI, organizations face several challenges. Data privacy and security continue to be top concerns, although their urgency has somewhat decreased due to the implementation of expanded certifications and governance frameworks such as SOC 2 and ISO 27001.

Talent shortages and insufficient AI literacy limit effective adoption, while budget constraints and the cost of high-quality AI solutions pose ongoing difficulties. Data quality is critical and poor inputs lead to flawed outputs. Additionally, factors such as regulatory considerations, organizational culture, and leadership buy-in greatly influence the success of implementation.

Effective adoption often requires top-down sponsorship and clear alignment with enterprise objectives. Transparent AI practices, where organizational data is kept under strict control, are essential for building trust and mitigating risk.

Beyond technical constraints, perception remains a significant barrier. Many organizations initially view AI as a universal solution when, in practice, it functions best as a supporting companion rather than a replacement for human expertise. While AI can assist with tasks such as drafting documents, analysing data and generating ideas, limitations remain in areas requiring judgement, creativity or contextual understanding. Improving organizational literacy around AI capabilities, expectations and risks is therefore essential to successful implementation.

Guardrails and responsible AI use

Organizations recognize the importance of establishing internal guardrails to ensure AI operates safely and effectively. Compliance with security and privacy standards, the inclusion of human-in-the-loop approvals, and the careful configuration of workflows help prevent autonomous agents from executing high-impact actions without oversight. These guardrails reflect ITIL’s guidance on AI governance, ensuring AI adoption aligns with organizational policies and risk frameworks.

Maintaining human oversight remains a cornerstone of effective AI integration. While autonomous agents can handle repetitive and predictable tasks, humans remain responsible for validation, trust-building, and decision-making. In incident management, AI can prepare recommendations, group alerts or generate problem investigation reports, but administrators must approve actions to prevent errors or misaligned outcomes. This human-in-the-loop governance ensures alignment with business objectives, builds user trust and provides continuous feedback to improve AI performance, particularly as organizations move toward greater levels of automation in high-stakes workflows.

Ensuring AI initiative success

A combination of the survey’s findings and customer experiences highlights four key areas essential for success: deliberate experimentation, aligning expectations with realistic outcomes, upskilling initiatives, and strong leadership support.

Starting with small, targeted experiments allows teams to learn without overcommitting resources. Establishing clear expectations helps prevent an overestimation of AI capabilities, while training and education across all levels enhance AI literacy and adoption. Active leadership involvement then ensures that these initiatives maintain momentum and align with strategic goals.

Companies like those mentioned in Forrester’s Total Economic Impact™ of Atlassian Jira Service Management study have shown measurable gains. For instance, service teams saved an average of 35 minutes per request, achieved $3.6 million in increased service desk productivity over three years and experienced a 30 per cent request deflection rate by using AI virtual agents.

Organizations must link experiments to meaningful enterprise metrics such as customer satisfaction, retention, and operational efficiency. Continuous measurement and feedback allow teams to refine implementations and demonstrate tangible value, particularly when AI investments involve significant upfront costs. Rather than deploying AI indiscriminately, successful organizations focus on targeted use cases tied to measurable outcomes and long-term business objectives.

Productivity, satisfaction, and the future

AI adoption is boosting both productivity and satisfaction, though it comes with its complexities. While employees benefit from increased efficiency and saved time, there are potential downsides if AI adds distractions or disrupts deep work. Survey results show that there is broad agreement on the benefits of AI, which improves data-driven insights, enhances productivity, and enables better prediction and prevention of service issues.

Customer service teams report the most significant gains, with 93 per cent acknowledging increased productivity and 91 per cent noting improved issue prediction. However, security and data quality remain important considerations, and measuring ROI continues to evolve as dashboards and metrics become more sophisticated.

Overall, AI adoption is enhancing workflows, supporting human creativity, and creating new opportunities. When combined with ITIL-aligned governance, clear objectives, and measurable outcomes, AI initiatives can deliver sustainable value while maintaining trust, compliance and operational resilience.