Building the AI-Proof MSP
Building the AI-Proof MSP
Time to read
6 minutes
Author
Nathan Marke, Chief Strategy Officer
at Giacom
Date published
August 2026
What does it take to become an AI-proof MSP? The answer is unlikely to be found simply by adding more AI products to the portfolio.
As featured in Comms Dealer, Nathan Marke, Chief Strategy Officer at Giacom, explores how AI could change the role of the MSP, from improving existing services to helping customers redesign processes, deploy AI agents and manage how work gets done.
Like many parents, I’ve found myself wondering what kind of work will still offer our children a secure future as AI develops. Every business should now be asking a version of that same question: how do we remain relevant and valuable as machines become capable of doing more of the work people do today?
For the channel, the conversation began with a focus on which AI products to sell. Increasingly, it is shifting towards what kind of business providers need to become.
We’re entering an intelligence supercycle. Mainframes automated transactions. PCs automated systems and processes. The internet, mobile and cloud transformed customer experiences. AI is beginning to automate thinking and decision-making.
AI will push MSPs from managing technology towards improving how their customers’ businesses work.
Doing the same things differently
Most AI activity in the channel today is focused on improving work that already exists. The service desk is an obvious example, with providers using automation to respond more quickly and handle a growing workload without adding people at the same rate.
AI can also change the economics of services that were previously difficult to provide profitably. Cybersecurity is a good example of this. A smaller MSP may not be able to employ all the specialist skills needed to provide an advanced security service, but AI-supported tools and automation could put more of that capability within reach.
These developments matter, but they’re unlikely to provide lasting differentiation. Every capable MSP will eventually use AI to improve its operations. Customers will come to expect it. The bigger opportunity lies in using AI to do different things.
From managing IT to managing work
I see the future MSP model developing across three layers.
The first is the technology foundation. Cloud infrastructure, connectivity and cybersecurity remain essential because every AI service will depend on what sits beneath it. As AI becomes more deeply involved in the customer’s operation, weaknesses in that foundation matter more.
The second is the use of AI to run those services better. Automation can improve support, identify problems earlier and give customers a clearer view of how their technology is performing. This should make providers more efficient, but it will increasingly become a basic requirement of running a competitive MSP.
The third begins when providers help customers redesign how work gets done. To do that, they need a deep understanding of the customer’s business, not simply the technology it uses. MSPs will need to recognise where AI can improve a process, where it should not be used and what must be in place for it to work safely.
An AI agent entering a business needs many of the same things as a new employee. It must be trained for the role, given access to the right information and kept within clear boundaries. Its performance must then be monitored, particularly when it begins making decisions or acting on the customer’s behalf.
The future MSP may not simply manage IT it may manage work. Helping businesses deploy agents and redesign workflows could become a major part of that role.
Become customer zero
The channel is still near the beginning of this journey. Some providers are making real progress, but the majority are not yet prepared for the scale of change ahead.
The best place to start is inside your own business. Before advising customers, MSPs should apply AI to their own operations. That means understanding where their information is held, improving its quality and identifying processes that can be automated safely.
Many providers will find that their data is spread across disconnected systems or held in a form that is difficult for either people or machines to use. Addressing that problem is valuable in itself, but it also builds experience that can later be applied to customers.
MSPs should be developing their understanding of data, security and business process now. These are likely to become more important than expertise in any single AI product.
What Giacom is doing
We‘re working through the same questions at Giacom. The first area is data, and we’re investing in creating a clearer and more consistent view of information across the business. This makes it more useful to our people today and better suited to agent-led processes in the future.
The second is orchestration. Our Cloud Market platform already automates much of the journey involved in buying, provisioning, supporting and billing for technology services. Those processes are currently designed mainly for people using a portal or connecting through conventional APIs.
Over time, they’ll also need to work for authorised AI agents acting on behalf of our partners. An agent should eventually be able to interact with Giacom securely and complete much of the work that requires a person to enter a portal today.
The third area is our people. As more of the administration surrounding technology is automated, our teams should be able to spend more time understanding providers’ businesses and helping them grow. This is likely to shift our go-to-market model away from helping partners complete transactions and towards helping them decide what to build and how to make money from it.
Of course, no one has all the answers. Over the coming months, we intend to learn from providers that are already building agent-led services and take that discussion further at Comms Vision.
The AI-proof MSP won’t simply add AI products to its portfolio. It will use AI to improve its own business while building the skills to help customers change theirs.
Read more
Watch Nathan’s webinar: How MSPs Can Monetise AI: Cashing in on Copilot
Download our Cashing in on Copilot Whitepaper