What is AITSM?
Initially, AITSM as an industry term appeared to represent the use of AI in ITSM—with both AI and ITSM sharing a single “I” for the sake of brevity. Oh, how we love our acronyms and initialisms in IT. Gartner presents AITSM with a slightly broader scope—as the application of AI, automation, and big data to ITSM to improve performance and reduce risk (mainly by eliminating human error). This is because of the close relationships that must exist between AI, automation, and big data to really make AI work in the world of ITSM. Practical AITSM must fit in to the ITSM people/process/tools/data/partners environment.
AI feeds on data—with the degree of success you get from it correlating with the volume of data you feed it. Of course, data quality counts too but “big data” inherently indicates significant volume. Velocity, variety, and veracity are the 3 other “V’s” of big data. The point here is that it is difficult to train an AI well on a small dataset.
AITSM: AI + Automation
AI-powered agent-assistance is a game-changer: augmenting human capabilities by having an AI mine ITSM data on the fly to instantly put the right information in front of the right people at the right time. Virtual agent assistants can effectively turn every agent into a “superhero” agent. An AI’s ability to join the dots between structured and unstructured data to present near-real-time insights is ideal for the rapid-fire service desk environment where agents are expected to support a broad and expanding array of customer issues.
But automation is where an AI comes out of the ivory tower and rolls up its sleeves. Instead of simply spending time poring over data and making suggestions for people to manually execute, enabling an AI with automation means it can pull levers in the IT environment to actually get things done. Meanwhile, people can get on with other things. Creative things. Transformative things. Human things.
AI + automation is how you achieve what we call “zero-touch IT operations”—like intelligent detect-and-correct scenarios where infrastructure issues/risks are automatically spotted and resolved at light-speed—and with no human intervention. Infrastructure is essentially self-healing. End user productivity is not disrupted. And the cost of these types of machine-solvable incident are reduced by 99% versus the human-driven detection/analysis/resolution process.
Another game-changing scenario is the end-to-end automation of responses to customer demands (incidents, requests, and queries from people, versus events from infrastructure monitoring). An AI-driven chatbot uses Natural Language Processing (NLP) and integration into ITSM toolset automation to not only interact with service consumers to answer questions—but to act on the issue to solve the problem and get them productive again. This could be anything from a simple password reset to provisioning access to a cloud app to setting up and testing a VPN connection.
This is the big opportunity of AITSM: Intelligent ITSM technology running a large part of IT service and IT operations management for you—so your people can devote more time to creating new innovations and providing great personal service to customers. For example, implementing a service desk chatbot will typically boost your support team’s capacity by 30%. Imagine what you could do with that.
Where do I start with AITSM?
The first question we often hear from customers is often “Where do we start?” As the science fiction writer Arthur C. Clarke once said “Any sufficiently advanced technology is indistinguishable from magic”. There is often a certain mystery that shrouds a paradigm shift, and AI is no exception. IT managers tasked with “doing AI” often experience a mix of emotions: a lot of excitement but also some fear and confusion.
Get Started
Contact us for a no-obligation chat with a consultant
Sign-up for blog notifications to make sure you don’t miss out
More About AITSM
Tell us what you think. Share your thoughts in the comments below.
ITIL 4© is a registered trademark of AXELOS Limited. All rights reserved.
