The story of historical digitalisation is one of traditional software systems. Software, as we once knew it, was a codified set of human rules: deterministic logic transcribed by humans from manual workflows and into digital, repeatable forms. The “old” software was rigid and literal, doing just exactly what it was told and no more.

Over the past 50 years, this approach has significantly mitigated internal systematic risks within businesses and delivered efficiency, predictability and scale to stakeholders.

Looking back, this historical digitalisation was largely the act of wrapping layers of software around incumbent value chains. It accelerated what enterprises already did but rarely reimagined what they did. Faster but not different. And importantly, sometimes freezing an enterprise’s “intelligence” to the moment the software systems were developed.

Generative AI (Gen AI) is very different. It is probabilistic and infers and reasons within ambiguity, rather than being bound by fixed rules programmed by humans. And it can adapt to new information without definitive instruction. Gen AI and its agentic AI systems are not hardwired — they can reason in new situations, generate new instructions and create new logic on the fly.

We have shifted from an era of encoding workflows and processes into software, so that they can be performed faster, into a world where Gen AI can entirely re-architect them. Instead of wrapping software around value chains, we are now able to embed intelligence into every part of a value chain through agentic AI.

With this, Gen AI allows us to reinvent what “work” is. When intelligence is universally available, every node in a value chain gains agency: reasoning, deciding and adapting in real time. Take procurement, which traditionally meant raising purchase orders based on forecasts, soliciting suppliers to respond to requests for quotes and then teams evaluating bids. In a world of agentic AI, an outcome-focused request to “deliver X units at Y specifications by Z date” can be delivered to AI agents that parse this into discrete parameters, then scan and evaluate supplier networks, simulate trade-offs and generate optimised options for sourcing autonomously.

Agentic AI delivers unprecedented problem-solving power far beyond the limited cognition and “tribal knowledge” crystallised in old software systems.

Gen AI’s impact is not about doing the same things faster. With agentic AI systems, we can change how we do things and redefine how we create value in business. Transforming the structure of business itself, rather than just the tempo of its existing routines.

Malaysia stands at the threshold of becoming a high-income economy. But as an open economy tightly integrated with global markets, the country faces a systemic risk from the increasing pace of Gen AI adoption worldwide.

Gen AI is poised to redefine the production function across entire economies, not just enhancing productivity but overhauling how businesses in these economies generate value. As intelligence becomes a utility, competitive advantage will accrue to those economies with businesses that absorb and utilise it fastest.

In previous technological shifts, catching up may have just required running faster on the same pathway. With Gen AI, being left behind means that others will have forged their own, different tracks. These new architectures of value creation will make catch-up exponentially more difficult.

Not keeping up could be the middle-income trap in 21st century form. There is a risk of being caught in it not for a lack of hard work, but because others have completely reinvented value creation, while the region is still focused on optimising old ways of working.

Hence, two imperatives emerge for Malaysia and the wider Asean region.

First, widespread Gen AI adoption in businesses across industries is essential to preserve or regain our competitive advantage. Second, Malaysia and the region must exploit the trove of globally distinctive data and domain expertise across energy, commodities, semiconductors and other high-growth verticals. Within these lie the raw materials for the region to create innovative agentic AI systems to navigate novel solutions.

This could be a defining moment for Malaysia and Southeast Asia. The region must seize the opportunity to not only mitigate the systemic risk of falling behind but also to lead the way in this next wave of digitalisation.

Embedding Gen AI into businesses in the region necessitates engaging with emergent AI innovation from the hotbeds of the US and China, where outsized investments in research and depth of talent are advancing AI technology fastest. However, unlike traditional software paradigms that were more amenable to “cut-and-paste” approaches, Gen AI demands nuance and context in application.

Private capital players with deep AI fluency and knowledge of business specificities in Southeast Asia are well placed to embed cutting-edge AI into regional businesses and manage the “donor rejection” risk inherent in imported innovation.

The region also needs seasoned innovation experts who can identify and help nurture new Gen AI innovations rooted in Southeast Asia’s unique data and domain expertise. These new ventures will require a deep grasp of the science behind Gen AI technologies alongside the commercial acumen to apply these from first principles in regional verticals.

Here, experienced investors who have successfully built Gen AI businesses can be changemakers in Southeast Asia. Those who have a track record in accelerating product-market fit in the AI space, and in judging what markets will (and will not) adopt, will be central in shaping the region’s first AI champions.

To be architects of the next phase of digital progress, Malaysia and the region must deftly translate deep AI innovation for regional adoption and nurture native AI champions. Private capital players are the catalysts that can bring these designs to reality.

Kamesh Raghavendra and Zubin Rada Krishnan are managing partners at The Hive Global AI Fund

Save by subscribing to us for your print and/or digital copy.

P/S: The Edge is also available on Apple's App Store and Android's Google Play.