AI proficiency has rapidly moved from “could have” to “need to have” for IT service providers (ITSPs) and MSPs. This is not about following a trend—it is about survival and relevance in the evolving technology landscape.
When I started AI Singapore in 2017, we struggled to find even 10 Singaporean AI engineers from 300 resumes. Today, we have trained over 400 AI engineers through our AI Apprenticeship Programme (AIAP), and the demand continues to grow exponentially. This transformation mirrors what ITSPs and MSPs must undergo.
Your customers are already asking about AI solutions. They want to know how AI can improve their operations, reduce costs and enhance customer experiences. If you cannot provide these answers or solutions, they will find vendors who can. We have seen this pattern across 1,000+ organizations we have engaged with through our 100 Experiments programme.
The shift is similar to when cloud computing emerged. ITSPs who adapted thrived; those who resisted risked becoming irrelevant. AI represents an even more fundamental shift because it transforms how we approach problem-solving across every industry vertical.
How to Get Started Building AI Skills and Solutions
Start with understanding, not coding. Many make the mistake of diving straight into Python programming or complex AI algorithms. Through our AI For Everyone (AI4E) programme, which has reached over 200,000 Singaporeans, we have learned that successful AI adoption begins with comprehension, not computation.
First, assess your organization's AI readiness using frameworks like our AI Readiness Index (AIRI). This 15-minute assessment across five pillars—organizational, ethics and governance, business value, data and infrastructure—will reveal where you stand and what gaps need to be addressed.
For skills development, follow this progression:
- 1. Foundation: Get your entire team AI-literate through programmes like AI for Everyone (AI4E)
- 2. Technical Depth: Identify champions who will pursue deeper technical skills
- 3. Hands-on Experience: Work on real projects, starting with simple automation use cases
We designed our AIAP programme around real-world projects because theory without practice is ineffective. ITSPs should similarly focus on solving actual client problems rather than building theoretical knowledge. Starting with projects that have clear ROI-document processing, workflow automation or predictive maintenance are good entry points.
Partner with organizations that have established AI programmes. We share our frameworks internationally, helping organizations build their own AI capabilities systematically.
The Biggest Challenge Most Companies Face
The biggest challenge is not technology, it is the gap between proof-of-concept and production deployment. Through our 300+ completed AI projects, we have observed that while many organizations can build impressive demos, few successfully deploy AI at scale.
This challenge manifests in several ways:
Infrastructure Readiness: Most organizations underestimate the infrastructure requirements. You need GPUs for training, high-performance storage and robust data pipelines. This is why infrastructure readiness is one of our five AIRI pillars. We learned this from our high-performance computing background—AI workloads demand serious computational resources.
Data Quality and Governance: Companies often claim they have data, but when projects begin, we discover it is scattered, inconsistent or inadequate. Our 100E programme now requires organizations to provide datasets upfront for baseline model development before project approval.
Talent and Culture: Building AI solutions requires a different mindset. It is iterative, experimental and often ambiguous. Traditional IT project management approaches do not work. You need teams comfortable with uncertainty and rapid pivoting.
End-to-End Pipeline Development: Many focus on model accuracy while ignoring deployment pipelines, monitoring, retraining mechanisms and integration with existing systems. Our platforms (AI/ML/Data Ops) team specifically addresses these aspects in every project.
The solution is to approach AI projects holistically. This is why our 100E programme includes not just AI engineers but also project managers, MLOps specialists and domain experts working together over 6 months to deliver production-ready solutions.
The Biggest AI Opportunity for ITSPs and MSPs
The biggest opportunity lies in becoming the bridge between AI technology and industry-specific applications. ITSPs and MSPs have something AI vendors lack—deep understanding of their clients' businesses, existing systems and pain points.
Based on our experience across sectors, here are the prime opportunities:
AI Integration Services: Most organizations need help integrating AI into existing workflows. They do not need another isolated AI tool; they need AI embedded into their current processes. ITSPs who can provide this integration will capture significant value.
Industry-Specific Solutions: Generic AI solutions rarely work. Through our 100E programme, we have seen how solutions for healthcare differ vastly from those for logistics or finance. ITSPs with vertical expertise can build tailored AI solutions that generic vendors cannot match.
AI Governance and Compliance: As AI regulations emerge globally, organizations need help implementing responsible AI practices. ITSPs can provide AI governance frameworks, bias testing and compliance solutions.
Managed AI Services: Many SMEs want AI benefits without building in-house capabilities. ITSPs can offer AI-as-a-Service, managing the entire AI lifecycle from data preparation to model monitoring.
Education and Enablement: There is massive demand for AI literacy and skills development. This need spans from C-suite executives to frontline workers. ITSPs can deliver customized AI education for their client base.
The key is to start where you have domain expertise. If you serve healthcare clients, focus on AI solutions for that sector. Build depth before breadth.
Advice for ITSPs
Develop a clear AI strategy before the market forces one upon you. The current AI vendor landscape resembles the dot-com era—lots of noise, inflated promises and genuine innovations mixed together. Here is how to navigate it:
Build Internal Competency First: You cannot evaluate what you do not understand. Invest in building internal AI literacy so your team can distinguish between genuine solutions and “AI washing.” Our AI4E programme was designed precisely for this purpose - to help organizations develop the foundational knowledge needed to make informed decisions.
Focus on Business Outcomes, Not Technology: When evaluating AI vendors, start with the business problem, not the technology. We have seen too many organizations buy impressive AI tools that solve no real business need. Use frameworks like AIRI to assess whether your clients are ready for specific AI solutions.
Create a Vetting Framework: Develop criteria for evaluating AI vendors:
- - Do they have proven deployments, not just demos?
- - Can they integrate with existing systems?
- - What is their approach to data privacy and AI governance?
- - Do they provide transparency in their AI models?
- - What ongoing support and model updating do they offer?
Start Small, Scale Smart: Resist the pressure to implement comprehensive AI transformations immediately. Our most successful 100E projects started with focused problems and scaled after proving value. Advise your clients similarly.
Partner Strategically: You do not need to build everything in-house. Partner with established AI programmes and vendors who complement your strengths.
Manage Customer Expectations: Be honest about AI limitations. It is not magic; it requires good data, clear use cases and organizational readiness. Use tools like AIRI to help clients understand their maturity level and set realistic expectations.
Remember, your value as an ITSP is not in having all the answers but in asking the right questions and guiding clients through their AI journey responsibly. The organizations that succeed will be those that approach AI as a capability to be built systematically, not a product to be purchased.
How can you adopt AI? Check out these top AI use cases.
Laurence Liew is director of AI innovation at AI Singapore. He will be speaking about Practical AI Adoption Strategies at the GTIA ASEAN Community meeting, 17 July in Singapore.
