The rate of adoption of artificial intelligence (AI)-based capabilities in IT service management (ITSM) appears to be outstripping any other practice- or technology-related ITSM change before it. As mentioned in our previous requests for AI survey help, this rapid rate of change means that a single annual AI survey, while helpful, is likely insufficient to track how organizations are using AI to improve their service management operations and outcomes.
Hence, we’re running our second annual AI survey just four months after our previous one. This repeats a similar AI survey from early 2025. This time, however, there’s a far greater focus on where the ITSM community is with Agentic AI rather than AI per se.
Agentic AI in The ITSM Spotlight
You can’t read about AI in ITSM these days without Agentic AI being mentioned. In the same way that workflow automation was a sought-after time- and cost-saving technology in the noughties, Agentic AI, or autonomous AI, is following “hot on the heels” of generative AI (GenAI) capabilities, or “assistive AI,” as the new way to improve ITSM operations and outcomes.
Low-, Mid-, and High-Agency AI Explained
There’s a lot of debate about what Agentic AI really is in ITSM right now. For example, discussions might dwell on whether there was “a human in the loop” at some point. For the purpose of this AI survey, I refer to the explanation given in the PeopleCert “AI in ITSM Tools” report, where an:
“…AI agent might align with one of the following agentic “levels”:
- Low-agency – making narrow decisions and taking simpler actions such as ticket categorisation, proposing resolutions, or running approved scripts. These AI agents augment human decision-making but might still be seen as acting autonomously.
- Mid-agency – extending beyond tactical support to handle multi-step reasoning and adaptive execution autonomously. Examples include diagnosing recurring issues, running more complex scripts, and providing automatic updates to stakeholders.
- High-agency – exhibiting goal-driven autonomy, including planning, reprioritising, and adapting outcome-focused strategies in real time. Examples include proactive problem detection, orchestrating cross-workflow solutions, and incident resolution with minimal human oversight.”
The key word for me, when calling something Agentic AI, is “autonomous.”
2025 AI Survey Key Findings
If you want to know what happens with the collated AI survey responses, the results of the 2025 AI survey can be found in the following three places:
- The full survey results are in a report here (an HCL Software report)
- An abridged version of these results is in an ITSM.tools article here
- The correlations that didn’t make the full survey report (or the associated ITSM.tools article) are in a second ITSM.tools article here.
Example AI Survey Insights
The following example data points and insights came from the 2025 version of this AI survey:
- 84% of survey respondents had a “positive” view of AI’s power in ITSM, while 13% selected more “negative” options.
- North American headquartered organizations were far more advanced in ITSM-tool-based AI adoption than those in Europe.
- Trust in AI had increased for 59% of respondents, while it had decreased for just 9%. Respondents in European headquartered organizations distrusted AI the most.
- Trust in AI was higher where paid-for AI capabilities (in ITSM tools) were used.
- 61% of survey respondents felt that corporate AI capabilities help them with their work, versus only 3% who don’t.
- North American headquartered organizations had seen three times the level of AI success in ITSM efficiency terms compared to those headquartered in Europe.
- 72% of survey respondents used free AI tools, and over three-fifths used them in addition to the available corporate AI capabilities.
- The top three achieved benefits of AI were cited as: Increasing employee productivity (32%), Better decision-making (20%), and Improving the end-user experience (20%).
- The level of Agentic AI adoption was similar across all organizational sizes. However, mid-sized organizations (101-500 and 501-2000 employees) were close to 67% more likely to have “No plans to adopt.”
- North American headquartered organizations, compared to their European counterparts, were over ten times as likely to have adopted self-healing capabilities.
Why Your Participation Matters
So please take this survey. The aggregated view of where individual organizations are with AI (and Agentic AI) adoption is helpful to the wider ITSM community and the progress made to date with this fast-moving technology area and ITSM opportunity.
Stephen Mann
Principal Analyst and Content Director at the ITSM-focused industry analyst firm ITSM.tools. Also an independent IT and IT service management marketing content creator, and a frequent blogger, writer, and presenter on the challenges and opportunities for IT service management professionals.
Previously held positions in IT research and analysis (at IT industry analyst firms Ovum and Forrester and the UK Post Office), IT service management consultancy, enterprise IT service desk and IT service management, IT asset management, innovation and creativity facilitation, project management, finance consultancy, internal audit, and product marketing for a SaaS IT service management technology vendor.
