AI in IT and Managed Services: The Industry That Built AI Is Being Changed by It

There's a particular irony in the IT industry being disrupted by AI. This is the industry that built the infrastructure AI runs on, employs the people who deploy it, and has been telling everyone else for years that technology transforms everything. Now the transformation is eating its own.

Managed Service Providers, IT support companies, system integrators, help desks — the entire ecosystem that keeps business technology running is being reshaped by the same AI they've been selling to their clients. And the people inside these organisations are experiencing what they've been helping other industries navigate: automation of their own work.

i find this both fascinating and slightly satisfying, if i'm honest. Not in a malicious way. But there's something instructive about watching the people who built the automation tools discover what it's like to be on the receiving end.

The MSP model under pressure

Managed Service Providers have a simple economic model. They manage your IT infrastructure, help desk, and technology operations for a monthly fee. The fee is partly based on headcount — the number of engineers and support staff dedicated to your account. More tickets, more users, more complexity means more staff means higher fees.

AI undermines this model directly. If AI can resolve 40-60% of help desk tickets without human intervention, the MSP needs fewer staff per client. Which means either the MSP charges less (lower revenue) or the MSP keeps the margin (but the client eventually notices and demands a price cut, or switches to a competitor who's already passed the savings on).

This is already happening. MSPs that were slow to adopt AI are losing clients to AI-forward competitors who can offer lower prices. MSPs that adopted AI early are seeing their per-client headcount drop, which is good for margins short term but means fewer billable staff long term.

The entire pricing model is shifting from "we provide X number of people to manage your IT" to "we guarantee Y outcomes using whatever combination of AI and humans is most efficient." The MSP that can deliver the same outcomes with fewer humans wins. The MSP that's still selling headcount is losing.

Tier-1 IT support: the obvious target

Just like in call centres, Tier-1 IT support is the most immediately affected. The help desk analyst who handles password resets, printer issues, VPN connections, software installations, and basic troubleshooting is doing work that AI handles increasingly well.

Modern AI-powered service desks can:

Resolve common issues automatically. Password resets, account unlocks, software provisioning, common error resolutions. These are handled by AI chatbots and automated workflows without any human involvement. The user describes their problem. The AI diagnoses it. The AI fixes it or walks the user through the fix.

Triage and route complex issues. When AI can't resolve the issue, it gathers diagnostic information, classifies the problem, and routes it to the right specialist with full context. The Tier-1 human used to do this triage. Now AI does it better because it has access to the full knowledge base and can correlate the reported symptoms with known issues instantly.

Provide 24/7 support. AI doesn't need shift patterns. It provides consistent support at 3am on a Sunday, which is when critical systems tend to break because the universe has a sense of humour.

The impact on Tier-1 headcount is significant. Some MSPs report reducing Tier-1 staff by 50-70% after deploying AI-powered service desk tools. The remaining Tier-1 staff handle the issues AI can't resolve and manage the AI system itself.

What's changing beyond the help desk

It's not just Tier-1 support. AI is working its way up the IT stack.

Monitoring and alerting. Traditional IT monitoring involved humans watching dashboards, reviewing alerts, and deciding what to investigate. AI-powered monitoring tools (AIOps) correlate alerts across multiple systems, identify root causes, suppress noise, and even trigger automated remediation. The operations engineer who used to sit in a Network Operations Centre watching screens is being replaced by AI that watches everything simultaneously and only involves humans when necessary.

Patch management and maintenance. AI can assess patch urgency, test compatibility, schedule deployment, and manage rollback if something goes wrong. The systems administrator who spent significant time managing patches and updates is seeing that work automated.

Security operations. AI-powered security tools handle threat detection, alert triage, and initial incident response. The security analyst role is evolving from monitoring to investigation and response (covered in more detail in the cybersecurity piece).

Cloud management. AI optimises cloud resource allocation, identifies cost savings, manages scaling, and flags security misconfigurations. The cloud engineer's role shifts from manual management to governance and architecture.

Infrastructure provisioning. Infrastructure-as-Code was already automating this, but AI is taking it further. AI can generate infrastructure configurations, optimise them, and deploy them with less human involvement. The infrastructure engineer who manually configured servers and networks is already largely a thing of the past; AI is now automating the code that automated the manual work.

The security opportunity

If there's a genuine growth area in IT services, it's security. As AI makes traditional IT support cheaper and more automated, the value proposition of MSPs is shifting towards security. Clients still need IT management but they're willing to pay less for it. Clients absolutely need security and they're willing to pay more for it because the threat landscape is expanding.

The Managed Security Service Provider (MSSP) model is growing. MSPs that can offer comprehensive security services — threat monitoring, incident response, compliance management, vulnerability assessment — are finding new revenue to replace declining support revenue.

This requires different skills. The help desk analyst who reset passwords is not easily retrained as a security analyst. But the experienced IT professional who understands systems, networks, and infrastructure has a foundation for moving into security. The skills gap is bridgeable, but it requires genuine investment in learning.

Cloud migration as a lifeline

The other growth area is cloud migration and management. Despite years of cloud adoption, a huge amount of business infrastructure is still on-premises. The complexity of migrating legacy systems to cloud environments, managing hybrid architectures, and optimising cloud costs creates work that AI assists but can't fully automate.

Cloud architects, migration specialists, and cloud optimisation consultants are in demand. The work is complex, context-dependent, and requires understanding both the legacy environment and the target cloud platform at a deep level. AI tools make these professionals more productive but don't replace the need for their expertise.

What the next-gen IT professional looks like

The IT professional of five years ago could build a career on technical depth in a single area. You could be a Windows server specialist, a network engineer, or a database administrator and have a stable, well-paid career managing and maintaining those systems.

The next-gen IT professional needs to be different:

Broader, not just deeper. Understanding how systems interconnect, how cloud and on-premises environments interact, and how AI tools fit into the overall architecture. The siloed specialist who only knows their one technology is vulnerable because AI can handle much of the routine work within any single domain.

AI-literate. Not necessarily building AI models, but understanding how AI tools work, how to deploy and manage them, how to evaluate their output, and how to troubleshoot when they fail. Every IT professional will be working with AI tools. The ones who understand them will manage and direct them. The ones who don't will be replaced by them.

Security-aware. Regardless of your primary specialism, security knowledge is becoming non-negotiable. Every IT system has security implications. Every AI deployment introduces new risks. The IT professional who can think about security implications across everything they do is more valuable than one who treats security as someone else's problem.

Business-oriented. The purely technical IT professional who communicates only in technical terms is less valuable than one who can translate technology into business outcomes. As AI handles more of the technical execution, the human value shifts towards understanding what the business needs and designing solutions that deliver it.

Automation-first. If your approach to IT problems is manual — log in, check things, make changes — you're working at a level that AI can replicate. The next-gen professional thinks in terms of automation, orchestration, and infrastructure-as-code. They solve problems once and automate the solution, rather than solving the same problem manually every time it recurs.

What to do if you work in IT or managed services

If you're in Tier-1 support: The trajectory is the same as call centres. Move up to Tier-2 or Tier-3, or move sideways into security, cloud, or automation. The Tier-1 help desk role has a limited shelf life.

If you're a generalist systems administrator: Specialise. The generalist sysadmin who did a bit of everything was valuable when each task required manual effort. AI handles the routine tasks across all those areas. You need to be deep enough in at least one area to handle the complex work that AI can't.

If you're at an MSP: Understand that your employer's business model is changing. If your MSP is slow to adopt AI, it's going to lose clients. If it adopts AI, it needs fewer staff. Either way, make sure your skills are aligned with where the industry is heading — security, cloud, automation — not where it was.

If you run an MSP: The brutal reality is that the Tier-1-heavy service desk model is not viable long term. You need to move towards outcome-based pricing, invest in AI tools, and reposition towards security and cloud services. The MSPs that make this transition will thrive. The ones that try to maintain the old model will be undercut by AI-forward competitors.

Watch for the pilot programme to restructuring pipeline at your company or your clients' companies. When a client starts piloting AI tools for their own internal IT, they're likely evaluating whether they still need an MSP for those functions. When your MSP starts piloting AI for service delivery, the headcount adjustment follows.

The one thing to do today: look at the AI-powered IT tools that are gaining traction in your specific area. Whether it's AIOps platforms, AI-powered service desks, or cloud management AI — learn them. The IT professionals who master these tools will manage and direct them. Everyone else will be managed out.