What does it take to build AI products in Africa? — Part 2
In my previous article, I shared an overview of the African AI landscape which is a growing market opportunity, and highlighted the importance of using market data as fuel to drive AI solutions to the local market. Being able to have qualitative market data will help you be more analytical and identify gaps in the market that can be filled by AI solutions.
When you have a clear understanding of the local market and a well-defined problem that you are ready to solve, you will need to consider some additional factors of importance that will determine what you build and with who you are building it.
Don’t rush your AI product development
I do agree that the initial launch of an AI product may have imperfections and that algorithms can still improve based on real data provided by user interactions. However, we must admit that it is crucial to ensure high accuracy and minimize associated risks before an AI model is made public for usage. You should know that most of the AI tools that have been built in a short time use pre-trained models, which can be effective and fast depending on the solution you want to build. Not all AI products have to be built from scratch.
In general, the development, training, and testing of AI products can take a considerable amount of time. Note that many of the successful and established AI products on the market are the outcome of extensive research and investment. This is one of the reasons why I will always encourage organizations to invest in research and spend the time needed to build their solution.
Don’t focus on trends, but build with purpose. Remember in the previous article, I mentioned that Africa doesn’t need fancy AI products but effective AI solutions. Instead of rushing to be the first to launch your AI product, it’s better to concentrate on making sure the product is accurate and efficient.
It’s also important to keep in mind that setting unrealistic deadlines will only put undue pressure on your team and create frustration around the product being developed. To ensure a successful deployment of your AI solution to the market, my recommendation is to take a hands-on approach. This involves using your time and energy efficiency by having the necessary resources and the right people onboard.
Embrace collaboration and work with the right people
A common belief is that in Africa we don’t have AI talents, but this is not entirely true. I agree that compared to some continents, we have not been exposed to the AI field very early on, and this is due to the lack of AI curriculum in our traditional programs, limited financial resources for higher education, especially in the field of AI, but this does not mean that there are no talented individuals in the continent who can contribute to the development of Artificial intelligence.
I and a lot of talented friends that I know in the AI space are the example of educated people who have developed their AI skills through a combination of formal and informal education and have been actively involved in the development and application of AI technologies to address local challenges and build innovative AI solutions.
The reality is that if you’re not willing to work with AI talents that don’t fit into certain predefined boxes, you’ll definitely have a hard time finding the right people here to work with. It is important to have an open-minded approach and engage with local talents with the expertise needed, who can commit and are willing to keep learning.
In general, AI skills are scarce and AI developers usually have to receive high salaries. As an entrepreneur, you have the option to establish an in-house team or outsource some of the skills required for your AI product(s) development. What you should have in mind if you are a newbie in the AI industry is, some of the essential skills you will need include that of data scientists, AI engineers, Data/AI product managers, domain experts, AI ethics experts, and lawyers. You may not have to get them all at once, but at some point in product development, you will definitely need each expertise listed to build effective products and build credibility in the local market.
It is essential to acknowledge that no matter how intelligent an individual may be, embarking on building an AI-driven product independently is unwise. Collaboration should be embraced, and I recommend working with a team with diverse backgrounds, skill sets, and gender, particularly including African-based talents when developing or expanding AI products in Africa. Understanding the local market is a critical aspect of building an AI-driven product in Africa. A strategy that involves local talents with practical experiences in the local market in the AI models development process, will provide an African market-centered approach, enabling your team to identify and reduce biases in the algorithms, and most importantly meet the AI products market fit.
