AI Appreciation Day is an opportune moment to take stock of how artificial intelligence is reshaping our lives. It is also a reminder that progress must stay transparent, responsible, and, above all, open.

Over the past decade, Southeast Asian businesses have enthusiastically adopted emerging technologies, and AI is no exception. The region’s real edge, however, lies in its appetite for open ecosystems. Open ways of AI development enable local innovators to build on global code, rather than starting from scratch, lowering entry barriers and accelerating time-to-market. These advantages are especially relevant in Southeast Asia, a diverse market where a “one size fits all” approach rarely works. Recent regional government discussions about AI – whether it’s Singapore’s landmark budget investment announcement, Malaysia’s newly launched AI Office, or Indonesia’s review of AI regulations – speak to the importance of AI adoption at a national policy level.

There is much potential to celebrate. Infrastructure built on open principles lets teams harness AI efficiently and ethically to deliver on the potential that AI promises. While closed, proprietary stacks can deliver quick performance wins, they often create legal, ethical and integration headaches. In contrast, open infrastructure helps organisations swap models, tooling and hardware as technology evolves.

The same foundations of open source also make it easier to embed guardrails, monitor performance and demonstrate compliance, which is crucial in regulated sectors such as financial services, healthcare and public sector functions. At the same time, the constant knowledge exchange baked into open source communities also creates a multiplier effect, enabling technology users to benefit from constant security updates while keeping costs manageable.

Practical applications of open source innovations in AI are already taking shape. For instance, an online retailer can adopt a virtual AI chat assistant onto its storefront in days, not months.

A hospital can run an in-house language model to keep patient data sovereign and audit-ready; and a factory can stream sensor data to lightweight edge models for real-time quality control, all while retaining the freedom to change components as better models or stricter rules emerge. Open infrastructure turns AI’s promise into transparent, ethical and adaptable reality while keeping user experience needs front and centre.

Ultimately, AI success is inseparable from architecture. Treat every layer – model, data and infrastructure – as interchangeable building blocks to avoid lock-in and stay agile as breakthroughs emerge.

On AI Appreciation Day we celebrate not just what AI can do, but how we choose to build it. With open source, businesses can innovate securely, transparently and on their own terms.