African Algorithms: Diversifying Artificial Intelligence

In 2015, Jacky Alcine, a software engineer in Oakland, California made a heinous discovery. While using Google Photos, he found that some of his friends were being identified by Google’s artificial intelligence system as gorillas. The tech and social justice worlds were shocked by the development. Google vowed to fix the problem immediately.

Image for post
Image for post
Alcine’s reaction to Google’s hiccup.

Some were not satisfied by Google’s response. They simply removed terms like “monkey”, “gorilla”, and “chimpanzee” from their API. Many would argue that this approach was just a quick fix that did not address the underlying problem of lack of diversity in the creation of artificial intelligence products.

Earlier this year, Google opened an artificial intelligence research lab in my hometown of Accra, Ghana. This center is the first of it’s kind in Africa. A development like this is essential to the prevention of incidents like Alcine’s. Moustapha Cisse of Senegal who will be heading the Research Center has experienced his fair share of obstacles in efforts to advance artificial intelligence on the continent

Image for post
Image for post
Moustapha Cisse, Head of Google A.I. Research Center in Accra, Ghana

At a 2016 AI conference in Barcelona, Spain that Cisse attended he was one of only 10 black people who attended the conference with 5,000 others. This prompted him to start Black in AI. An organization that is comprised of 1,000 researchers, students, and AI enthusiasts who are committed to striving for a better and more inclusive AI environment for black people. Despite Black in AI’s great success and large support Cisse has been unable to attend many conferences where his research was being presented because of issues acquiring a visa to travel from the African continent.

Although there may be a myriad of issues that plague the space of artificial intelligence for Africans, there is growing optimism that there can be a quick change in the tide of fortunes for the continent. Last year, venture capitalists poured 560 million dollars of investments into African tech startups. Google is advising more than 60 startups through a program they started called Startup Launchpad Africa. According to the International Monetary Fund, six of the ten countries with the fastest growing economies is African.

There is clearly an opportunity for advancement. Fortunately there are some Africans that are taking on the challenge of shifting the paradigm. Below I have highlighted some companies that are taking on the task of changing the narrative:

Image for post
Image for post

Founded in 2016, Nigerian startup Kudi has raised $5.9 million in funding that started with a seed round from Y Combinator, the accelerator that brought us Airbnb, Dropbox, and Coinbase. Kudi has developed a chatbot for bill payments and money transfers that operates through Facebook Messenger and Skype. The chatbot uses phone networks instead of data connections since only 39% of Nigeria’s population has internet access.

Image for post
Image for post

Given all of Africa’s farmlands, it’s no surprise that our next two startups were recently highlighted in an article on “10 AgTech Startups for Agriculture in Africa.” Founded in 2014, Cape Town, South Africa startup Aerobotics has raised $4.8 million to develop a geospatial intelligence platform that enables pest and disease detection in tree crops using drones, satellite imagery, and of course artificial intelligence to interpret all the big data that’s being collected.

Image for post
Image for post

The home of our next startup is the northernmost country in Africa and a place where Star Wars fans congregate to see famous set locations used during the earlier films. Founded in 2016, Tunisian startup Datavora has raised $893,000 to develop a market data platform for e-commerce. The startup gathers information like product assortment, prices, and specifications on tens of thousands of SKUs available online to provide benchmarks and competitive data for e-commerce providers. The algorithms consolidate and structure the data, match similar products, analyze market trends, and predict market behavior. The tool is used by retailers and brands to build sales and marketing strategies for ecommerce. Datavora covers more than 2,000 retailers in 50 countries and is updated multiple times a day. Prices start at $400 a month for monitoring one product category in three countries.

The narrative is changing one day at a time. In order for there to be a more inclusive world, there must be more African involvement in artificial intelligence. With the development of these new companies, gaffes like Google’s and the obstacles that Cisse faced will be a thing of the past in the near future.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store