by Tan Aik Keong (AK)
At this year's NVIDIA GTC 2026, Jensen Huang took the stage once again in his signature black leather jacket. The keynote wasn't just a product launch — it read more like a manifesto about the future: AI is no longer just a tool. It's becoming infrastructure, the way electricity and the internet did, reshaping the global economy and society along the way.
Huang opened with a key claim: the AI Factory will become the core asset of the enterprise of the future. Businesses used to rely on data centres to process information; in future, they'll build infrastructure purpose-built to produce "intelligence" itself. These AI factories, through training and inference, continuously generate models, insights and automation capability — becoming the actual source of enterprise competitiveness.
That concept is particularly worth Malaysia's attention as a developing digital economy. Are we ready to move from "using AI" to "producing AI"? Across Southeast Asia, most businesses are still stuck at the application layer — chatbots, e-commerce recommendation engines. Huang's keynote is a reminder that the real value will concentrate at the layer underneath: control over compute, models and data.
NVIDIA also unveiled its next-generation GPU architecture at the event, reinforcing its dominance in AI compute. From training large language models to real-time inference, these chips are faster, more power-efficient, and deeply optimised for generative AI. In other words, the AI race ahead will, to a large degree, still be a hardware race.
That raises a real question: where does a country like Malaysia sit in this compute race? Large-scale AI training is currently still heavily concentrated in the US and China. Without local compute infrastructure, we risk staying dependent on external platforms long-term — a cost issue, but also a question of data sovereignty and technological self-determination.
Another major theme of the keynote was agentic AI. Huang demonstrated AI that no longer just passively answers questions, but actively executes tasks — writing code, designing products, managing business processes. These AI agents will operate more like digital employees, lifting productivity significantly.
But that raises its own question: as AI takes on more and more cognitive work, how do we define the human role? In Malaysia, plenty of middle-class jobs — customer service, admin, parts of professional services — face potential disruption. AI brings efficiency gains, but it may also widen employment-structure imbalances.
Worth noting: Huang repeatedly emphasised the "partner ecosystem" throughout the keynote. He pointed out that industries from cloud computing to automotive manufacturing are all partnering with NVIDIA to build AI applications together. That ecosystem strategy means the competition ahead won't just be between individual companies — it'll be between platforms and ecosystems.
For local businesses, that's both an opportunity and a challenge. Joining a global AI ecosystem can bring rapid access to capability — but without a core value of your own, you risk becoming a mere appendage.
Huang also walked through AI's promise in healthcare, robotics and autonomous driving — from virtual drug discovery to smart factories, genuinely exciting scenarios. But for an ordinary reader, the more grounded question is: when do these technologies actually land? And will they benefit every level of society?
Looking back at the whole keynote, Huang didn't offer simple answers — he posed a bigger question: in the AI era, are we participants, or spectators?
For Malaysia, the key issue isn't whether to embrace AI — it's how to position ourselves. Do we become consumers of AI technology, or gradually build our own capability? Do we simply follow the global giants, or carve out differentiated strength in specific areas?
Perhaps the real question worth sitting with isn't whether AI will change the world — it's whether we're ready, now that the world already has.
In the wave NVIDIA is leading, the future won't wait for the hesitant. The question is whether we're willing to step into this transformation ourselves.
Part of the AK AI Corner column. Originally published in Oriental Daily (东方日报) on Apr 4, 2026.