Developing and Launching a Successful AI Product in Africa: Lessons from KETI

Recently, I was privileged to gain early access to KETI, an AI-powered medical chatbot developed by Chil AI, a company founded by Ugandan entrepreneur Shamim nabuuma . This experience opened my eyes to the immense potential of AI in addressing Africa's unique challenges, particularly in healthcare.

KETI lets users obtain medical interpretations for complex diagnoses and prescriptions on WhatsApp. Users ask about medical terms or upload medical documents (test results or prescriptions), and KETI responds with clarity.

What's more? it translates into several languages, including some of Africa's most prominent ones. KETI's special quality lies not just in the complexity or accuracy of its responses, but in its simplicity and ease of access.

As I reflect on my experience with KETI (which you can try by saving the number +256 742 006671 on WhatsApp), I've begun to research what it truly means to build a solid AI solution with actual relevance in our continent. Here's what I've learned:

1. Identify a Pressing Need

KETI addresses a critical issue in Africa: the lack of accessible medical information. Many Africans struggle to understand complex medical terminology or interpret test results, especially in their native languages. By focusing on this specific need, Chil AI has created a product with immediate, tangible benefits.

Practice:

• Conduct extensive field research in various African communities.
• Engage in conversations with potential users from diverse backgrounds.
• Create a list of common challenges that could potentially be addressed by AI.

2. Simplify the Complex

KETI's genius lies in its simplicity. It takes the complex world of medical jargon and makes it accessible through a platform most Africans already use daily: WhatsApp. This approach eliminates barriers to entry and ensures widespread adoption.

Practice:

• Break down your complex AI solution into simple, user-friendly interfaces.
• Use familiar platforms or technologies that your target audience already understands.
• Eliminate technical jargon in your user communications; focus on benefits and outcomes.
• Develop clear, step-by-step guides for using your AI product.
• Create visual aids or infographics that explain your product's functionality simply.

3. Build with Local Context in Mind

By offering translations into prominent African languages, KETI demonstrates a deep understanding of its target market. It's not just about providing information; it's about delivering it in a culturally and linguistically relevant way.

Practice:

• Conduct extensive market research to understand the specific needs and challenges of your target African markets.
• Incorporate local languages and dialects into your AI model's training data.
• Consider local cultural norms and practices when designing your user interface and experience.
• Adapt your AI solution to work with existing infrastructure limitations (e.g., low-bandwidth areas).
• Engage local experts in your field to ensure your solution is contextually appropriate.

4. Prioritise Ethical AI Development

In a continent where AI regulation is still developing, ethical AI development becomes a personal and professional responsibility. It requires a commitment to global best practices and a deep understanding of local implications.

Practice:

• Educate yourself and your team on global AI ethics standards and best practices.
• Develop a clear, written ethical framework for your AI development process.
• Implement rigorous testing for bias in your AI models, considering local cultural factors.
• Establish transparency in your AI decision-making processes, making them easily explainable to users.
• Create an ethics advisory board that includes diverse local perspectives.
• Regularly audit your AI's performance and decisions for ethical considerations.
• Prioritise data privacy and security, even in the absence of strict local regulations.

5. Collaborate with Local Institutions

Partnerships with local universities, tech hubs, and government institutions can provide valuable resources, expertise, and credibility. These collaborations can also help in scaling your product across different regions.

Practice:

• Reach out to local universities for research partnerships.
• Engage with tech hubs and incubators for mentorship and networking opportunities.
• Explore government initiatives that support AI development in your target sector.

6. Design for Scalability and Sustainability

While your initial focus might be on a specific community or region, design your product with scalability in mind. Also, consider the long-term sustainability of your solution – how will it continue to operate and evolve?

Practice:

• Develop a modular architecture that can be easily adapted to different contexts.
• Create a sustainable business model that doesn't rely solely on donor funding.
• Plan for continuous improvement and updates based on user feedback and changing needs.

7. Invest in User Education and Support

Launching the product is just the beginning. Invest heavily in user education and support. This might mean partnering with local community leaders, creating easy-to-understand user guides, or setting up local support centres.

Practice:

• Develop a comprehensive user education programme, including workshops and training sessions.
• Create multilingual support materials in various formats (text, audio, video).
• Establish a network of local ambassadors who can provide on-the-ground support and feedback.

8. Measure and Communicate Impact

To ensure continued adoption and support, it's crucial to measure and effectively communicate the impact of your product. This isn't just about numbers – it's about real stories of how your AI solution is improving lives.

Practice:

• Develop clear, relevant metrics for measuring impact.
• Collect and share user testimonials and success stories.
• Regularly publish impact reports that are accessible to both stakeholders and users.

9. Iterate and Improve

Remember that launching your product is not the end of the journey. Continuously gather feedback, monitor performance, and be ready to iterate and improve your solution.

Practice:

• Implement robust feedback mechanisms, including in-app surveys and community feedback sessions.
• Regularly analyse user data to identify areas for improvement.
• Stay updated on advancements in AI and assess how they can enhance your product.

KETI serves as an inspiring example of how AI can be leveraged to address critical needs in Africa. It demonstrates that successful AI products in our continent don't necessarily need to be the most technologically advanced. Instead, they need to be the most relevant, accessible, and impactful.

As we continue to develop AI solutions for Africa, let's take inspiration from KETI's approach. Let's focus on creating products that seamlessly integrate into people's lives, address real needs, and bridge critical information gaps. By doing so, we can harness the power of AI to improve millions of lives across our continent, one WhatsApp message at a time.

Remember, our goal isn't to create the most advanced AI – it's to create the most impactful AI. Let's focus on solutions that truly matter to our communities, even if they don't fully understand the complexity behind them. After all, the best technology is the one that seamlessly improves lives without drawing attention to itself.