Matt and I attended the same high school. He studied mechatronics engineering in college. Today, he shares his first essay on Atoms & Bits.
-
Many people across the continent now use mobile phones, computers, and the internet in their everyday lives. All this has been possible through leapfrogging.
As for AI, that is another story.
AI stands for artificial intelligence, which is the simulation of human intelligence processes by machines or computer systems. AI can mimic human abilities like communication, learning, and decision-making.
Afrika’s love for AI is growing rapidly. Every week, there is a summit, a conference, and podcasts where technology enthusiasts talk about how AI will revolutionize the continent.
They use the word disruption to show how this technology will revolutionize healthcare, fintech, insurance, agriculture, and their all-time favorite, education. These views are simplistic but not realistic.
Tesla uses the technology to achieve full self-driving (FSD) and in its humanoid robots. Microsoft integrated LLM into Copilot, and OpenAI just signed a deal to bring ChatGPT (NLP) across Apple Intelligence later this year.
Adobe, as the name suggests, uses generative AI in Photoshop. Nvidia has become the world's most valuable company by selling GPUs that machine learning models are trained on.
A large language model (LLM) is an AI model that has been trained on large amounts of text so that it can understand language and generate human-like text.
Generative AI is a type of technology that uses AI to create content, including text, video, code, and images. A generative AI system is trained using large amounts of data so that it can find patterns for generating new content.
Additional applications include recommendation systems, targeted web advertising, virtual assistants (Siri, Alexa, and Google Assistant), chatbots, spam filtering, language translation, speech recognition, facial recognition and image labeling, precision farming, crop and soil monitoring, automated machinery, fraud detection, algorithmic trading, music, robotics, natural language processing (NLP), and deep fakes.
Natural language processing (NLP) is a type of AI that enables computers to understand spoken and written human language. NLP enables features like text and speech recognition on devices.
Top universities in Afrika are not focused on research and development in AI. There are not enough funds allocated in that area. AI is now being included in the syllabus as a subject to learn for exams but not as a potential area for R&D.
If you look at MIT and other prestigious universities, a clear divide between them and Afrikan universities can be seen. How are you going to win the AI race if your top institutions are focused on gender studies?
There is no AI development without electricity. Only 50% of the population has access to electricity. 5 out of 10 people live in the dark, as most of the time there’s load shedding. Computational power for AI comes from electricity.
Data centers require constant power supply. As more models will need to be developed, more consumption of electricity will be needed, and a 50% electricity penetration rate won’t cut it; you can’t run GPUs on solar power or electricity from generators.
AI requires internet connectivity. As of June 2022, the internet penetration rate was 43.2% across the whole continent. Training models require large amounts of data, most of which is online. AI requires distributed computing, which is possible through cloud computing, and accessing the cloud requires the internet.
Better internet connectivity ensures models are updated and deployed on time. If internet access is costly and unreliable, it discourages AI development. How are you going to train machine learning models if most of your data is in file cabinets (offline) dusting away on shelves?
In Afrika, the backbone of the continent is agriculture, trade, and commodities. How often does an Afrikan man come across a problem in his daily life that would be solved with AI? It is nearly nonexistent.
You could argue and say that in banking, healthcare, and education, I would counter you and say, How many people have access to banks? How many people have access to proper imaging and diagnosis?
How are you going to talk about AI in education if the primary schools you graduated from nearly 20 years ago don’t have computers, internet connectivity, or even enough desks?
Governments set vision, frameworks, and policies that govern innovation. In the USA, DARPA (Defense Advanced Research Projects Agency), NSF (National Science Foundation), and NIH (National Institutes of Health) have received funds for AI development.
Some initiatives were set as early as the 1950s to lead us to the silicon age of AI, but in Afrika, this had never occurred. The U.S. leverages its strong private sector, research institutions, and strategic initiatives to drive AI innovation, but this is not seen in Afrika.
The tale of AI development in Afrika could end the same way crypto did. Crypto looked so promising; it was to bring transparent government spending, easy remittances, and central bank digital currencies, but they are all in vain.
In the end, crypto didn't change Afrika as much as people had hoped. For Afrika to really benefit and keep up AI, the continent needs much better basic infrastructure, electricity, internet, and overall systems.