When I first started reporting on the IT industry in the late 1990’s, the big shift everyone was talking about was the rise of services and by the early 2000’s it was cloud that typified as-a-service mantra. Vendors, resellers, and MSPs were racing to redefine themselves as cloud-centric and IT service providers, swapping capital investment for operational agility and expenditure. It was a transformation that reshaped the entire channel. But the rise of artificial intelligence feels different — more urgent, more disruptive, and with the potential to be even more transformative than cloud ever was.

The new platform shift

AI isn’t just another technology trend sitting alongside automation or analytics. It is fast becoming the platform on which the next era of IT will be built. McKinsey’s 2025 Technology Trends Outlook notes that the firm has now rolled all of its previous AI-related trends into a single overarching category because it underpins almost everything else happening in technology. For MSPs, that is a clear signal that this isn’t an optional add-on or niche experiment, but a fundamental change in how software and services are delivered.

The impact can already be felt across every segment of the market. Whether it is Microsoft embedding copilots into Microsoft 365 or AWS integrating generative models into its cloud infrastructure, AI is no longer a future consideration. It is here, it is commercial, and it is rapidly becoming integral to how businesses operate. As one Gartner analyst put it earlier this year, “AI is shifting from a technology capability to a business imperative.”

The uncomfortable truth about automation

Yet there is a side to this transformation that few in the technology industry want to confront directly. AI isn’t only about boosting productivity or enabling smarter decisions, it is also about displacement. Unlike previous revolutions that automated physical labour, AI has the potential to automate cognitive labour — the work of knowledge professionals who once considered themselves insulated from machine competition.

Research by the Brookings Institution suggests that over 30 percent of all job hours could be affected by AI-driven automation. The Pew Research Center estimates that nearly one in five workers in developed economies are in roles most exposed to AI replacement. Early adopters are already using it to trim costs, cut human-error rates, and compress project timelines. For business leaders, the economic attraction is obvious: AI promises not just productivity gains but reductions in the wage bill.

For MSPs, that matters on two fronts. First, clients will expect to see the efficiency gains that AI enables reflected in pricing and service models. Second, the labour-intensive nature of many managed services — monitoring, ticket handling, first-line support — means parts of the traditional MSP model could soon be automated away if providers fail to evolve.

Challengers on the attack

AI is also redrawing competitive boundaries. The same dynamic that allowed cloud to unseat traditional infrastructure providers is now playing out again, only faster. New players are using AI to attack long-established incumbents who once seemed unassailable.

In search, for instance, OpenAI’s new ChatGPT Atlas browser has blurred the line between web navigation and conversational search, allowing users to explore the internet through natural dialogue rather than typed queries — a direct challenge to both Google Search and Microsoft’s Bing.

In enterprise software, open-source AI models such as Meta’s Llama are eroding the lock-in once held by hyperscalers. Even cybersecurity, one of the most entrenched segments of the channel, is being reshaped. A wave of AI-driven detection and response vendors is challenging the likes of Palo Alto Networks and Cisco with systems that promise faster analysis and fewer human interventions.

The lesson for MSPs is clear: no vendor relationship is guaranteed, and no technology stack is immune. AI is enabling new market entrants to rewrite the rules, often with remarkable speed. The partners who monitor these shifts closely, and are ready to pivot as they gain traction, will be better positioned to adapt when the landscape changes.

The goldrush problem

Of course, the scale of investment pouring into AI has created its own distortions. The current wave feels like a digital goldrush, with venture capital flooding into thousands of startups promising to revolutionise everything from customer support to software development. But as anyone who remembers the early days of public cloud will recall, not everyone survives.

During the first decade of cloud computing, dozens of providers promised to challenge Amazon Web Services. Today, only AWS, Microsoft, and Google have achieved global scale. The AI sector is likely to follow a similar pattern. Many companies are exciting right now, but few have proven business models or sustainable differentiation, and most will falter before they reach profitability.

Despite the hype, AI’s return on investment remains difficult to quantify. PwC’s most recent AI Predictions report highlights that most enterprises are still “struggling to link pilot projects to measurable business outcomes.” For MSPs trying to help customers justify expenditure, that is a familiar story. Cloud adoption only accelerated once clear ROI metrics emerged — the savings between running a data centre as capital expenditure and consuming cloud infrastructure as a predictable operational cost. AI has yet to deliver an equivalent benchmark.

A double-edged sword for MSPs

So where does this leave managed service providers? Somewhere between excitement and anxiety. On one hand, AI can transform operations. Tools that automatically generate reports, triage tickets, or predict hardware failures can increase margins and reduce costs. On the other, those same capabilities are available to every competitor, including a new wave of AI-first service providers who are leaner, hungrier, and free of legacy systems.

The danger lies in complacency. Believing that AI will arrive fully formed one day, ready to integrate neatly into existing workflows, is a mistake. The shift is already underway, driven by customers who expect their suppliers to be using these technologies internally. MSPs that fail to articulate an AI strategy risk appearing outdated long before they realise it.

Proving the value

To bridge that credibility gap, the industry needs more verified, quantified case studies showing how AI improves outcomes. Without evidence of tangible ROI, it is all too easy for AI to be dismissed as hype. The challenge is that those success stories require rigorous measurement of time saved, cost avoided, revenue generated, and customer satisfaction improved — and few organisations are yet tracking those metrics consistently.

MSPs are uniquely positioned to change that. They sit close enough to customer operations to measure impact, yet have the technical expertise to understand how AI tools are deployed. If they can demonstrate results with the same clarity that cloud providers once used to show the CAPEX-versus-OPEX advantage, they will help move the market from curiosity to conviction.

A force for good, but not for everyone

It would be wrong to view AI purely as a threat. Properly harnessed, it will drive enormous efficiency, open new service categories, and help address the IT skills shortage that so many MSPs face. Yet it will not be painless. Some suppliers, distributors, and intermediaries will struggle to find relevance if they cannot integrate AI into their offerings. The same technological force that enables one MSP to thrive could quietly erode another’s margins.

As history shows, technology revolutions rarely lift everyone equally. The winners are those who experiment early, learn fast, and adapt their business model before necessity forces their hand.

Building literacy before it’s too late

So what should MSP leaders do now? The first step is education. Building AI literacy across the organisation, starting with the senior leadership team, is essential. Without a baseline understanding of what AI can and cannot do, it is impossible to make informed strategic decisions. That knowledge needs to flow down through teams so every manager and technician can recognise where AI might improve efficiency or unlock new value for clients.

The second step is to start experimenting internally. Use AI to automate routine administrative work, generate internal reports, or enhance customer communications. Experience is the best teacher, and those lessons will be invaluable when advising clients.

Above all, treat AI not as a product to resell but as a capability to master. Businesses that treat it as a long-term competency rather than a short-term add-on will be the ones still standing when the hype fades.

Because one thing is certain: AI will reshape the managed services landscape just as cloud once did, perhaps more so. The question for every MSP is whether they intend to ride that tsunami, or be swept away by it.