Why Your SME Should Pay Attention to a Nonprofit AI Race
You are running a profitable business in Malaysia. Your customers speak Bahasa Melayu, Mandarin, and Tamil. You are trying to integrate AI tools to handle customer service, generate marketing content, or automate your orders. But the AI stumbles on the cultural context of your Hari Raya ads and defaults to English whenever it gets confused.
This gap is costing you trust, time, and market share. It is exactly the problem a nonprofit called Current AI is racing to fix. Backed by $400 million from governments, foundations, and tech labs, it is building what it calls “the World Wide Web of AI”—a free, open, and public alternative to the Big Tech systems we currently rely on (TechCrunch).
TL;DR: A well-funded nonprofit is building open-source AI infrastructure that prioritizes language diversity and community data control. For an SME owner in Malaysia, this signals a future where you are not locked into Silicon Valley’s roadmap, and your customers’ local languages actually work inside the tools you adopt.
The Problem with English-Only AI for Your Malaysian Market
Every major AI system today—OpenAI, Google, Anthropic—belongs to a private company. This means their priorities follow their biggest markets. English is the default. Other languages are treated as optional add-ons, if they are considered at all.
Current AI CEO Ayah Bdeir put it bluntly in a recent interview: “Half the world’s spoken languages face extinction. And with English driving the largest language models and AI systems, a bulk of the world’s languages and, consequently, cultures and communities are left behind” (TechCrunch).
For a Malaysian business owner, this is not an abstract problem. If your AI assistant cannot handle the mix of languages your customers actually speak at the pasar malam or in your WhatsApp group, it is not truly serving your business. You are adapting your workflows to the tool, rather than the tool adapting to you.
“If AI is truly a transformative technology, if it’s going to change every aspect of everyone’s life, there has to be a public alternative. Like the World Wide Web, available to anyone, for free.” — Ayah Bdeir, CEO, Current AI
What Is Current AI Actually Building?
Current AI is not just talking about a better future. It is shipping hardware and software that prove the model works.
Offline Devices for Underrepresented Languages
At the India AI Summit, Current AI teamed up with the government’s Bhashini language division to create Suno Sutra, a pocket-sized offline device that runs AI in 22 Indian languages with no internet required (TechCrunch). The device is open-sourced, meaning anyone can take the blueprint and improve it or adapt it for their own language community.
An Open-Source Chatbot Built by a Coalition
Earlier this month at the AI for Good Summit in Geneva, Current AI launched Alpha Chat, an open-source chatbot assembled in just seven weeks by ten organizations including Hugging Face, Mozilla, and the MIT Media Lab (TechCrunch). Each contributor brought a piece of the stack—a language model, safety tooling, and computing power—and the result is free for anyone to use or extend.
Grassroots Grants Putting Community First
Current AI allocated $3.2 million in its first grant round to four organizations across Kenya, Lebanon, and the Brazilian Amazon (TechCrunch). The projects include building AI datasets across 50 African languages for health and farming, digitizing Arab cultural history, and creating offline AI tools for Indigenous Amazon communities—all with strict data consent protocols that let communities halt the process at any point.
“There are different models and proposals for who owns data in various communities,” Bdeir told TechCrunch, “but one thing is sure: it shouldn’t be a company in Silicon Valley trying to make a select few thousand people wealthier” (TechCrunch).
Closed AI vs. Open AI: A Business Owner’s Comparison
To make this concrete for your decision-making, here is how the two approaches stack up for an SME owner in Malaysia.
| Factor | Closed / Big Tech AI | Open / Public AI (Current AI Model) |
|---|---|---|
| Data Control | Your business interactions train a global model you cannot see or stop. | Data is stored locally or within your community. Consent is designed into the pipeline. |
| Language & Culture | Optimized for English. Local languages are an afterthought, if they are included at all. | Built for diversity from the ground up. 22 Indian languages is only the starting point. |
| Customization | You can only use the interface and features they decide to offer. | Fully open source. You can adapt the model for your specific industry and install it on your own terms. |
| Long-Term Independence | High switching costs. You are dependent on a single company’s roadmap and priorities. | Modular and open. You are not locked into one provider. The infrastructure belongs to the public. |
It feels like the open model gives you significantly more flexibility as a business owner. You are not just a user of the technology; you can influence where it goes and how it serves your specific customers.