AI takes ITSM from simple automation to a predictive engine for CIOs. This article deciphers the keys to this transformation.
Artificial intelligence in ITSM: from an automation lever to an engine for transforming IT services
By Isabelle Roth, Practice lead ESM Europe at OpenText
Artificial intelligence (AI) is gradually establishing itself as a structuring component of IT service management (ITSM). After numerous questions about the available tooling and initial experiments with integrations of external solutions, it is now integrated into the heart of modern ITSM platforms. This development responds to double pressure: on the one hand, increased user expectations in terms of quality of service and speed of processing and, on the other, a security issue in the use of AI.
The challenge for IT departments is no longer whether AI has its place in ITSM, but how to exploit it in a pragmatic, controlled and value-generating manner in a sustainable manner.
Automate without dehumanizing
One of the first contributions of AI to ITSM lies in the intelligent automation of tasks. Incident, request and change management has historically relied on highly structured, but often cumbersome and time-consuming workflows. Thanks to natural language processing (NLP) and machine learning capabilities, AI now makes it possible to automatically categorize tickets, change their priority according to user feedback, suggest experts to accelerate resolution, etc.
This automation is not intended to replace IT teams, but to relieve them of repetitive or low value-added tasks. By reducing the time spent analyzing content or searching for information, teams can focus on resolving or processing their tickets.
A profoundly transformed user experience
Modern ITSM can no longer be thought of solely from the perspective of IT teams. User experience has become a key performance indicator. In this area, AI plays a decisive role by facilitating access to services and streamlining interactions between requesters and technical teams.
Virtual assistants or conversational agents, when well designed, allow users to formulate their requests in natural language, without having to navigate catalogs often perceived as complex. They do not only point to documents or offers but respond to the user’s search and guide them in their efforts. However, the success of these systems depends on the ease of implementation and the quality of the responses. The AI must suggest the most relevant documents and offers in the context of the requester. By being permanently available, it strengthens the relationship between IT and users.
When instructions are not enough, automation associated with support or service requests then makes it possible to reduce response times while improving the perception of the service provided.
Knowledge management enhanced by AI
Knowledge management is another area where AI brings significant value. AI cannot work without quality data, but maintaining an up-to-date, relevant and actually used knowledge base is an ongoing challenge for organizations. AI can help automate the creation of content and titles for knowledge articles, drawing on a complex set of data from tickets. AI can also generate recommendations to improve the quality of existing articles.
AI can suggest solutions to agents by identifying the most relevant knowledge articles but also requests, incidents, changes, etc., thus improving resolution times.
By analyzing unstructured data from requests and incidents, AI also makes it possible to identify the most important themes and thus enrich the knowledge base or service catalog or generate problems for recurring incidents.
Conditions for success: data, governance and skills
Despite its promise, AI in ITSM is not a silver bullet. Its performance depends closely on the quality of the data available and the settings put in place. We must therefore first identify the data that the AI models or agents will consume: internal or external, and check their security. A public AI solution can be very rich but provide you with answers that go against your rules. Your business data can also be made accessible to everyone. Your internal data may be incomplete, obsolete or unsuitable for consumption by an AI, thus considerably limiting the expected benefits. Implementing solid data governance is therefore an essential prerequisite.
Generally speaking, the adoption of AI raises issues of skills and change management. IT teams must be trained not only in using the tools, but also in understanding how they work. The transparency of models, the ability to adapt and enrich models like AI agents are important for administrators.
Finally, human acceptance remains a key factor. AI must be seen as an aid to agents and not as a replacement threat. Clear communication on the contributions of AI to their daily tasks and team involvement from the design phases are essential to promote buy-in.
Towards a more strategic ITSM
AI introduces a profound change: the transition from reactive logic to predictive logic. By analyzing large volumes of data, algorithms can identify trends and take existing data into account to improve the quality of process processing.
These new capabilities transform the way teams approach process management and service continuity. By combining automation, prediction and contextual intelligence, AI offers CIOs the opportunity to strengthen the perceived value of IT and better align services with business challenges.
Provided it is deployed with discernment, AI strengthens ITSM, helps it to structure itself and places it sustainably in a logic of value creation.
