The Time the App Said 'Not Yet' — And How I Learned to Listen to the Silence
When building AI-powered tools for everyday life in Kenya, I've learned that not every problem has an immediate answer. Sometimes, the most valuable lessons come from the moments when the AI says 'Not yet' — and what happens next.
Building AI-powered products in Kenya is a journey of trial and error. One of the most humbling experiences came during the development of Local Dialect, the language learning platform for Kenyan and African languages. I was excited about the idea of using AI to help users learn regional dialects, but I quickly ran into a wall — the AI wasn't responding the way I expected it to. It wasn't giving me the right answers, and it wasn't helping me structure the lessons the way I wanted. I kept asking it the same questions, and it kept saying 'Not yet.'
At first, I felt defeated. I had assumed that the AI would be able to understand the nuances of Kenyan languages and help me design an engaging learning experience. But I was wrong. The AI wasn't failing me — it was showing me that I needed to rethink my approach. I had to go back to the drawing board and figure out how to structure the content in a way that the AI could actually work with. It was a moment of clarity, but also a moment of frustration.
What changed was my willingness to listen — not just to the AI, but to the people who would be using the app. I started talking to language teachers, learners, and community members to understand what they needed. I realized that the AI wasn't the only piece of the puzzle. The content, the structure, and the context were just as important. I had to make sure that the AI was working with real, human language — not just data.
This experience taught me that AI is a tool, not a magic solution. It can help, but it can't replace the human element. When I built Local Dialect, I had to balance the power of AI with the realities of language learning. I had to be patient, I had to be flexible, and I had to be willing to listen — even when the AI said 'Not yet.'
If you're building something and it's not working the way you expect, you're not alone. I've been there, and I know how frustrating it can be. But sometimes, the most valuable lessons come from the moments when the AI doesn't give you the answer you want. It's in those moments that you have to dig deeper, ask better questions, and find new ways to solve the problem. That's how we build real, useful tools — not just for ourselves, but for the people who need them most.
Join the newsletter for more stories like this. Every Monday and Friday, I share insights on building AI-powered tools that make a real difference in everyday life. Subscribe here: https://wambugumartin.com
