How to build AI capability locally, before it’s too late
How to build AI capability locally, before it’s too late July 10, 2026 What’s new: In a recent Asia CEO catch-up held in Singapore, members shared a common frustration. They want to do more with AI but are not always sure how to start. Why it matters: Regional CEOs are not waiting because they doubt AI. Their hesitation stems from a lack of authority, budget, or technical confidence. Commonly, Asia teams are told they must wait for centralised direction from HQ. Yes, but waiting carries its own cost. Asian competitors are moving fast to experiment with AI. By the time HQ ships a solution, the use cases and tech stack may miss the mark in the local market. What Asia CEOs are doing: Below are ways MNCs in Asia are pressing ahead while staying within company-mandated compliance and governance frameworks. The Asia region has a deep, technically savvy AI ecosystem. Western MNCs operating here can take advantage of the best of both worlds. 1. Build AI literacy team-wide Several CEOs admitted they did not fully understand AI themselves. Instead, they focused on exposing teams to the technology and creating space for experimentation. A leading global tech firm made AI training mandatory for everyone, setting a common baseline across the company. Three years ago, our entire company went through a global AI training — it was like a day-to-day AI course. Everyone had to do it, regardless of their position or knowledge. It was so good I wanted my kids to take it. But most consumer and industrial firms have been slower to develop global AI training programs. Asia CEOs know their job is to motivate their teams. But where to start? The Asia CEO of a food wholesaler and retailer turned to a software vendor for assistance in running an off-site AI workshop for her team. I often joke with IT that I am a dinosaur. I know I don’t have the answers, but I wanted to inspire our team. We asked our software supplier, Microsoft, to organise a one-day workshop to show our people what is possible with AI. This spurred a lot of conversations. If we don’t have a goal, then we have to go out and find it. Too often, we are internally focused. Workshops can be a useful way to bring functions together and cross-pollinate ideas. Silos block digital transformation; the same goes for AI. But a one-off event fades without reporting lines and incentives behind it. Inertia returns unless change management is deliberate. For the Asia CEO of a luxury brand, post-workshop leadership requires adjusting the org chart and KPIs. I agree it is very difficult to know where to start as a CEO without a technical background. We also held a one-day workshop to motivate people to start with AI. We ensured people came from different functions. But this was just the beginning. There needs to be constant focus and revised reporting lines to keep an initiative going. Inculcating a growth mindset is more essential than ever in these turbulent times. AI does not stand still. Neither does customer innovation. In many cases, AI adoption is more of a change-management and people problem than a technology one. I encourage a growth mindset in our company. If you don’t have a growth mindset, then people close themselves off and nothing changes. We are a very dynamic organisation. Our advantage is that we are central to our customers’ business. Either we adapt, or we’re out. We are very project-driven, always opening new locations or changing our networks. The work is constantly changing. It is not like a bank where the office is the same, and you replicate it. This has made us quite open to AI. (Logistics) 2. Create AI champions While AI literacy matters, enterprise-wide mandates rarely inspire the passion and drive to make big changes. A tech CEO found it effective to start with a small AI-savvy team to develop edge cases. About six years ago, I set up a small data team of nine people that became the genesis of our AI learning in the region. This team must work within our company’s data compliance policies, and we partner with IT for funding. But they report directly to me, not to IT. The best ideas are not always centralised; they come from the edge. It takes a flexible organisation that empowers people to be innovative. Another firm gave AI access to its most eager volunteers first, regardless of title or seniority. Those volunteers are set to become AI ambassadors and change agents for their teams. When we launched Copilot, instead of meting out licenses to leaders, we asked for AI volunteers. In my region, Asia, we had about 200 licenses for 900 people. We can see the most active users are young, move more quickly than the rest, and are highly motivated to figure things out for themselves. We asked how we can push this group further. So we decided to make the most active users AI ambassadors and leverage their passion. 3. Identify your use cases Two themes emerged on where to start with use cases. First, target the workflows where staff spend the most time. The largest surface area is where AI delivers the best ROI. It may be that the most high-impact AI opportunities are hidden within processes executives thought were already automated. Second, let the people accountable for business results drive the work. IT should not own AI transformations, but it can be essential in supporting them. The Asia CEO of a hardware firm shared how they used focus groups to surface new ideas. We start by narrowing down to one or two areas that can bring the most efficiency and used internal focus groups to get more people involved. We looked at where employees spend most of their time. Admin paperwork turned out to be massive, which was expected. More surprising was how much time sales spent preparing quotes for customers. We had thought that
