Hard pill to swallow:
AI will steal Service Desk jobs.
But it will also make room for a new breed of Service Desk employees.
How will AI replace humans in the IT process where the most human interaction happens? What benefits will it generate for IT managers and businesses from AI at Service Desk? What new positions will appear as a result of the implementation of AI at the Service Desk?
Interaction with users
Quality of interaction with users at the Service Desk defines the visibility of the whole IT Department. Doing it wrong kills the professional image of even the best technical experts. People hate to talk with chatbots. When they do, they only strive to hack it to allow them to speak with an actual human. But it will change soon.
Chatbots are getting better and better at faking human language. ChatGPT is already good enough to have pleasant conversations with people. AI evolves insanely fast. Soon it will be good enough to put it in front of the users, and it will be more helpful than 80% of Service Desk technicians. Users will prefer to talk to chatbots than real Service Desk technicians.
Identifying issues
For many types of issues, AI doesn’t have to wait for users to report problems. With suitable monitoring tools, it can identify issues before users notify it. This mechanism can be built into the operating systems and not even bother the Service Desk. With Microsoft investing billions in OpenAI, we can expect it to use the AI potential in many ways. One of them can be self-diagnosis built into Windows OS.
Dispatch of tickets
Service Desk technicians dispatch tickets to the second and third support line teams. Large organizations have many support teams, which differ in expertise, location, and other details. Human Service Desk technicians find it overwhelming to decide which team to dispatch the ticket to.
AI is brilliant at processing large data sets. Trained well, it will dispatch tickets faster than humans and with better accuracy. It can use the learned logic to decide which team to assign to the ticket. But it can also use a large set of static rules and procedures.
Fixing software issues
The more issues Service Desk solves by itself, the quicker user gets the requested service. Solving IT issues doesn’t seem so evident for AI to take away from human technicians. From a human perspective, it’s often complex and hard to automate.
You start fixing the issue with a proper diagnosis. As with identifying issues, AI can analyze many symptoms and log messages to find the root cause of the case. With the ability to recognize context, AI should do an excellent job at this step.
Having the root cause figured out, AI can find a way to fix it. How can it learn to find solutions? You can teach it by rewarding right solutions and penalizing wrong. You can also give it static rules to follow. You’ll get the best results if you use both methods. The current level of ChatGPT’s ability to recognize context seems enough to find the right answers when fine-tuned.
AI can already write scripts in various programming languages. It can program features based on descriptions in natural language. Nothing more is needed to fix software issues. AI can execute a script it developed remotely. There are already many tools to run scripts, even on many endpoints simultaneously.
Fixing hardware issues
Service Desk usually doesn’t fix hardware issues. Still, as it is one of the operations needed for users by the IT teams, let’s consider this area as well.
Fixing hardware issues seems like more of a challenge for AI. It would need a kind of robot to perform complicated manipulations to replace computer parts. The robot would need maintenance and service itself. Human interaction will still be necessary in this way or another.
On the other hand, even at home, we tend to replace devices instead of repairing them. In the business environment, there can be a stock of devices with pre-defined hardware configurations. AI can configure a personalized user environment on the new device within a few minutes. Users provided with a new computer and the environment they are used to can get back to work quickly. And the broken device can be sent to hardware-service-as-a-service to repair or recycle.
Even in specialized hardware service, technicians won’t have to diagnose the issue and decide what to replace. AI can do most of this work for them and provide them with step-by-step instructions on what to do.
Verification
Solutions, especially those provided automatically, need verification. The same algorithms which identify issues can verify if everything is OK after applying a solution. It can be done automatically, but a well-trained chatbot can also interview users after the Service Desk intervention pleasantly and efficiently. If AI identifies an issue as not solved, it can redirect it to the solving algorithm again or… hand it over to a human specialist.
Documentation
People hate filling out ticket forms with all the details of the support process. AI doesn’t hate. It will fill in all the needed information and free human technicians from this struggle. Nuance DAX is an AI-based service that can fill medical documentation. DAX does it only by listening to conversations between doctors and patients. A similar AI tool should do well with the documentation of IT issues.
Will AI replace Service Desk technicians?
Well… Yes, but not all of them. At least for some time, we will need Service Desk employees. But they will solve only more complicated, rare issues where AI won’t be able to help.
Still, there will be a lot of work for Service Desk AI trainers. Someone will have to fine-tune AI to suit specific business environments. The more customized environment, the more training AI will need.
Unfortunately, Service Desk will become less entry-level employees friendly. That’s a pity as it is an excellent forge for future IT systems administrators and IT managers.
The good news is that the AI trainer position seems to fit well with the Gen-Z mindset. Gen-Z employees hate doing things themselves. They are used to a reality where technology does their dirty work. They prefer to operate the high-tech solution rather than do the actual job themselves. It seems that AI makes the world becoming more Gen-Z friendly! At least the Service Desk…