Microsoft has unveiled a groundbreaking inclusive artificial intelligence initiative in Kenya aimed at making advanced technology more accessible and relevant to communities often underserved in global innovation efforts.
Known as Project Gecko, the program brings together Microsoft Research Africa in Nairobi, Microsoft Research India, the Microsoft Research Accelerator team in the United States, implementation partner Digital Green, and a network of academic and philanthropic supporters. The initiative seeks to design AI systems that reflect real-world needs in emerging markets, rather than relying on tools built primarily for English-speaking regions with strong connectivity and high digital literacy.
A central innovation of Project Gecko is the Multi Modal Critical Thinking Agent (MMCTAgent). This AI model allows people to connect intuitively without the need for sophisticated technical abilities since it can read a variety of input formats, including audio, photos, and video. By making the technology available on open-source platforms, Microsoft promotes a cooperative approach to inclusive AI development by encouraging researchers worldwide to enhance, modify, and use the system.
One of the first sectors targeted by the rollout is agriculture, a critical industry for millions of Kenyans and the wider continent. Due to their restricted internet connectivity and reliance on oral instruction in regional languages, many rural areas find it challenging to utilize traditional AI techniques. Project Gecko expands upon Digital Green’s current farmer support platform, which makes use of voice-activated help and a sizable knowledge base with thousands of training videos in many regional languages. It used to be difficult for users to navigate this library and extract pertinent information.
Farmers can use Project Gecko to ask questions in their native tongues and get voice, text, or video answers. Even direct timestamps are provided by the system, directing users to specific instructions for the activities at hand. Widely spoken Kenyan languages like Swahili, Kikuyu, Kalenjin, Dholuo, Maa, and Somali are currently supported by the program. The accuracy of the model has been improved through the use of thousands of hours of speech data gathered from Kenyan users. Additionally, more than 100 farmers have actively tested the system, offering feedback that resulted in improvements like a clearer question-and-answer flow and more useful advice that is in line with local farming conditions.
Beyond agriculture, the broader goal of Project Gecko is to create a framework for inclusive AI design. In order to ensure that emerging markets gain equally from technology breakthroughs, Microsoft plans to issue a comprehensive guide that will assist developers in creating AI systems that are useable across a variety of locations, languages, and connectivity circumstances. The initiative supports Microsoft’s other initiatives in Kenya, such as the AI skilling program aimed at 1 million people by 2027 and the construction of a green data center to house an Azure Cloud region for East Africa, by combining skilling programs, infrastructure investment, and useful AI applications.
Project Gecko underscores Microsoft’s commitment to inclusive innovation, demonstrating how AI can be tailored to local contexts while supporting education, economic development, and sustainability. By creating tools that are culturally and linguistically relevant, Microsoft hopes to empower communities across Kenya to engage with AI in meaningful ways, setting a precedent for inclusive technology initiatives in Africa and beyond.
