AI has moved from curiosity to business priority. Across Asia Pacific, the conversation is shifting. Business leaders are no longer asking, “Can we do AI?” but rather, “Is it delivering?” The novelty is fading, and the expectations are getting sharper. What’s the return? What value are we unlocking? What problems are we actually solving?
But there’s a blind spot in many of these conversations. While companies rush to adopt AI tools and add compute power, few are spending enough time fixing what truly matters – the data feeding those systems.
Here’s the reality: messy data means messy AI. If your organisation has five versions of the same dataset floating around, sitting in different departments, and no one’s quite sure which is the most accurate or up to date, you don’t get intelligence – you get noise. And in sectors like finance, healthcare, or government, that noise can lead to dangerous decisions. Throwing more compute at the problem won’t help – it’ll just get you to the wrong answer faster.
To get AI right, we need to do the hard, unglamorous work: cleaning up data, standardising it, making sure everyone’s working off the same version of the truth. That means breaking down silos, setting clear rules on how data is handled, and investing in the right systems to track and manage it.
This isn’t a one-off task – it’s a mindset shift. Think of it like a flywheel: clean data leads to better models, better models help make smarter decisions, and those decisions generate even more useful data. But the flywheel only spins if the input is solid.
We also need to take data accountability seriously. That means knowing where your data came from, how it’s been changed, and who has access to it. Without that transparency, we can’t build systems we trust and neither can regulators, customers, or the public.
On AI Appreciation Day, let’s shift the focus to the foundation. Because real progress doesn’t come from adopting the latest tool, but from building the right infrastructure, mindset, and culture to support it.
