The Mozilla Foundation recently announced its “Rise25: Leading the Next Wave of AI” list, highlighting 25 individuals, and I am honored to be among them.

The selected individuals share a commitment to “equity,” working across various fields with the spirit of open source to address issues such as racial, gender, and economic disparities. Whether in technology or business applications, the AI solutions proposed by these honorees are grounded in this principle.

Open-source platforms are like public libraries—where everyone is equal. Utilizing the open-source model is crucial to countering the risk of generative AI being monopolized by tech giants and to promoting resource sharing.

Why is this important? Because as AI models grow larger and require more data, akin to having a bigger brain and reading more books, the cost of training also increases, often becoming affordable only to a few tech companies. By open-sourcing foundational models, various sectors can combine their own data and customize AI systems according to their needs, creating solutions that better meet real-world demands.

For example, this year’s Rise25 honoree, Peter Gault, emphasizes close collaboration with educators through his open-source initiative, Quill.org. Gault’s platform uses AI to provide free resources for students to improve their reading and writing skills, while also offering high-quality curriculum content to teachers, giving students with limited resources greater learning opportunities.

Quill brings together a group of K-12 teachers to create the content they need. Quill’s Teacher Advisory Council (TAC), with over 300 members, tests every learning activity before it is launched, building a vast sample of real classroom scenarios to guide AI development. This approach ensures that AI-generated feedback is relevant to classroom needs, making it a practical tool. In contrast, many generative AI systems led by technical teams often provide direct answers when students struggle, failing to effectively address students’ learning needs.

Currently, Quill is limited to English usage. Fortunately, many in Taiwan are also dedicating their efforts to open-source models, datasets, and evaluation systems. A great example of this is the “TAiwan Mixture of Experts (Project TAME),” a collaborative effort between industry experts from chemical engineering, electronics, healthcare, content creation, and law, working with a team from National Taiwan University.

In the education sector, projects like the Taiwan Knowledge Base (ATK) open-source initiative cover a wide range of topics including history, culture, politics, education, and current events. These projects are curated and validated by educators and subject matter experts, ensuring that AI tools align with local contexts and are more accessible to the public.

As discussions continue on whether generative AI is merely a passing trend, those committed to open source are gradually tackling previously unsolvable problems, laying the groundwork for a more equitable and inclusive future.