AI and IT Support: What's Actually Happening and What to Do

The honest assessment

If you work in IT support — helpdesk, service desk, desktop support, whatever your organisation calls it — you've probably already noticed the ground shifting beneath your feet. AI isn't approaching IT support. It's already embedded in it.

The tier-1 IT support function, the one that handles password resets, software installation requests, basic troubleshooting, and "have you tried turning it off and on again" calls, is being automated aggressively. Microsoft's own data shows that its AI-powered IT support tools can resolve up to 65% of common IT issues without human intervention. ServiceNow's virtual agent handles millions of IT support interactions. Freshservice, Jira Service Management, and Zendesk all have AI capabilities that can diagnose issues, suggest solutions, and even execute fixes automatically. Password resets — which have historically accounted for 20-30% of helpdesk ticket volume — are increasingly handled by self-service portals with AI-powered identity verification.

The broader numbers are clear. Gartner predicts that by 2027, chatbots will become the primary customer service channel for roughly a quarter of organisations. For internal IT support specifically, the trajectory is even more advanced because the environment is more controlled and standardised than external customer service. Internal IT systems are known. The user base is defined. The problem types are predictable. This is exactly the kind of structured environment where AI excels.

But there's another side to this. The complexity of IT environments is increasing, not decreasing. Cloud migrations, hybrid working infrastructure, cybersecurity threats, regulatory compliance, and the explosion of SaaS applications mean that IT departments have more complex problems to solve than ever. The simple stuff is going to AI. The hard stuff is getting harder. If you can move from the simple stuff to the hard stuff, your career isn't threatened — it's potentially accelerated.

Your exposure level: High

High. Tier-1 and tier-2 IT support roles are among the most automatable functions in any organisation.

The fundamental reason is that IT support is inherently a triage-and-resolution workflow. User reports problem. System categorises problem. Known solution is applied. Ticket is closed. This workflow is the definition of what AI handles well. The ITIL framework that most IT support organisations follow was essentially designed to make the process as standardised and repeatable as possible. AI doesn't struggle with standardised and repeatable. It excels at it.

The financial incentive for organisations is significant. An average IT support ticket costs £15-£22 to resolve with a human agent. AI-resolved tickets cost a fraction of that. When IT budgets are under pressure — and they always are — automating tier-1 support is an obvious efficiency measure. The justification is usually framed as "freeing up the team for more complex work," and to be fair, that's sometimes genuinely what happens. Other times, it's a headcount reduction exercise.

Where the exposure drops is in roles that involve physical infrastructure, complex troubleshooting, project work, and strategic IT decisions. Setting up a new office network. Migrating a company to a new cloud platform. Investigating an intermittent issue that affects some users but not others and doesn't match any known pattern. Managing vendor relationships and IT procurement. These require the kind of contextual understanding, physical presence, and creative problem-solving that AI can't replicate.

The 90-day action plan

1. This week: learn your ITSM platform's AI features inside out. Whether you use ServiceNow, Jira Service Management, Freshservice, or something else, dive into every AI and automation feature available. Auto-categorisation. Virtual agents. Knowledge base suggestions. Automated workflows. Understanding these tools is essential because they're the ones that are automating your current tasks. You want to be the person who configures them, not the person they replace.

2. Week two: start tracking the tickets AI can't solve. Keep a log of the complex issues you handle — the ones that require investigation, creativity, or physical presence. Network issues that don't match documented problems. Software conflicts between applications. Hardware failures that need diagnosis. This log is evidence of your value beyond automation.

3. By day 30: pursue a cloud certification. AWS Cloud Practitioner, Azure Fundamentals, or Google Cloud Digital Leader. These are entry-level cloud certifications that are achievable in a month of focused study. Cloud infrastructure management is growing while traditional desktop support is shrinking. This is the most direct path from "IT support" to "cloud engineer" or "systems administrator."

4. By day 45: develop scripting and automation skills. Learn PowerShell, Python, or Bash scripting. The ability to write scripts that automate complex IT tasks is what separates a support technician from a systems engineer. Use AI tools like ChatGPT and Claude to help you learn — they're excellent at explaining code and debugging scripts. Every script you write that automates a complex process demonstrates engineering ability.

5. By day 60: explore cybersecurity fundamentals. Security is the fastest-growing area of IT, and it's one of the least automatable. Study for CompTIA Security+ or look into CyberFirst programmes if you're in the UK. IT support experience gives you a solid foundation for security roles because you understand systems, networks, and how users actually interact with technology (usually badly).

6. By day 75: build expertise in a growth area. Identity and access management. Microsoft 365 administration and security. Network architecture. DevOps. Infrastructure as code. Pick one area and go deep. The IT support generalist is the most vulnerable role. The specialist with deep knowledge in a specific, growing domain is in demand.

7. By day 90: have the career progression conversation. "I've noticed that AI is handling more of our tier-1 tickets, which is efficient. I've been developing my skills in [cloud / security / automation / networking] and I'd like to move into a role that focuses on the more complex infrastructure and security work. Here's what I've done to prepare and here's where I think I can add the most value." Don't wait for your role to be restructured. Get ahead of it.

AI tools you should be using this week

• ChatGPT for Work — Your fastest tool for troubleshooting unfamiliar issues. "Users are getting error 0x80070005 when trying to access a shared folder on Server 2019. What are the most likely causes?" ChatGPT can also help you write scripts, create documentation, draft knowledge base articles, and explain complex technical concepts to non-technical users who need to understand what happened.

• Claude for Work — Excellent for working through complex, multi-step troubleshooting. Describe the symptoms, the environment, and what you've already tried. Claude's methodical approach is well-suited to IT diagnosis. Also strong for reviewing scripts and configuration files for errors. When you're debugging a PowerShell script at 4pm on a Friday, Claude is a patient colleague who won't judge you.

• Microsoft Copilot for Work — If you're in a Microsoft environment (and statistically, you probably are), Copilot integrates with the admin tools you use daily. It can help with Intune management, Azure AD queries, Exchange Online troubleshooting, and generating reports for management. The admin centre integration is particularly useful for common management tasks.

• Zapier AI for Automation — Connect your IT tools to create automated workflows. When a new starter is added to HR system, automatically create accounts, assign licences, send welcome documentation, and schedule equipment setup. IT onboarding and offboarding automation saves hours per employee and reduces the errors that come with manual processes.

What to say in meetings

When management discusses automating the helpdesk: "I'm supportive. The routine tickets are the least interesting part of the job, frankly. What i'd like to discuss is how we reinvest the capacity. We have infrastructure projects, security improvements, and cloud migration work that's been waiting for resource. AI handling the password resets means we can focus on the work that actually improves the organisation's IT capability."

If colleagues are worried about helpdesk automation: "The tier-1 stuff is going to be automated. That's just reality. But look at everything else that needs doing — security, cloud, automation, the projects that keep getting deprioritised because we're too busy fixing printers. There's more complex IT work than there are people to do it. The shift is an opportunity if we develop the skills."

In performance reviews: "I've resolved [X] complex tickets this quarter and started developing skills in [cloud/security/automation]. I've also written [Y] knowledge base articles and automated [Z] processes. I'd like to discuss a development path toward [systems engineering / security / cloud administration]."

If the worst happens

If you're made redundant from an IT support role, your technical foundation transfers to many adjacent roles. You understand networks, operating systems, hardware, software, and user behaviour. You can troubleshoot systematically. You're used to working under pressure with multiple competing priorities. These skills apply to systems administration, network engineering, cloud operations, cybersecurity, DevOps, and technical project management.

Adjacent roles to consider: junior systems administrator, cloud support engineer (AWS, Azure, and Google Cloud all have entry-level support roles that value IT support experience), junior security analyst, technical account manager at a tech vendor, IT trainer, or technical writer. Many IT support professionals also find opportunities in managed service providers (MSPs) where the work is more varied and hands-on than in large corporate helpdesks.

Here's the thing i want you to understand. IT support is a foundation, not a ceiling. Some of the best systems engineers, security professionals, and IT leaders i know started on a helpdesk. The helpdesk taught them how technology actually works when real users are involved, which is different from how it works in a textbook. That experience is genuinely valuable. The role as it exists today is being automated. The career path it leads to is wide open. Make the move while you still have the luxury of doing it on your own terms.