The emergence of cheap or free AI tools is being eagerly embraced by those with smartphones and the ability to get online. As governments and legislators struggle to get their heads around the implications of this powerful technology and work out how to bring in regulations for its safe use, millions of people are enjoying its ability to save time, helping them to transforming raw data into essays, exam answers, or, with a bit more work, even videos and podcasts.
Even in developing countries where electricity and internet access is limited (it’s estimated that over 570 million people in Africa lack electricity), there is enthusiasm for the potential of AI. In the Democratic Republic of the Congo (DRC), for example, a nation riven by internal conflict, poverty and vast inequality, educators are seeing the impact of AI.
“It is obvious that our country is lagging behind in terms of new technologies for one reason or another,” says Benjamin Sivanzire, a teacher in Beni, North Kivu Province. “Many parts of the DRC do not even have traditional methods of communication, or even radio or television.”
Benjamin Sivanzire, a teacher in Eastern DRC.
However, even though Mr. Sivanzire and his students are not yet able to make use of AI in their classes, they are seeing it being used in the wider culture, often in a negative way, to manipulate public opinion.
The teacher underlines the importance of educating people to distinguish between verifiable information and lies. “There are videos created by artificial intelligence that show images that are not real and have been created for propaganda purposes,” he explains.
Bursting the Silicon Valley bubble
One concern that is frequently raised is the extent to which the development of AI tools is concentrated in the hands of a relatively small group of people.
Farida Shahid, the independent Special Rapporteur on the right to education, shares these concerns.
“AI algorithms are being made by individuals who often sit in a particular location, such as Silicon Valley, where the people who make and test them have their own biases,” she says. “Often the algorithms don’t do well at recognizing people with dark skin. They also have great problems with people who are autistic and don’t like looking into cameras.
“Another example is the UK where, recently, an AI programme was used to grade exam papers. This led to decisions that were biased against people from certain ethnic backgrounds. We really need to look at this issue more closely, starting with the human rights perspective, and I think that’s where the U.N. role comes in: if you increasingly rely on AI as the source of verification, you’re going to have problems because you are using a framework which privileges white males, and doesn’t reflect the whole gamut of people’s lives and experiences”.
The urgent need to expand the developer talent base has been identified by the UN as central to ensuring that a wide variety of voices are heard in the “EdTech” (educational technology) space.
Shafika Isaacs, the head of technology and AI at the UN agency for science, technology and education (UNESCO), says that the number of African EdTech startups has been mushrooming in recent years, with entrepreneurs experimenting with the AI-enabled digital tools which could support learning and teaching across many different contexts, including in African languages, and local dialects.
“I’ve personally engaged with a startup that matches high school students to career pathways, including choosing the right university, community college or even entrepreneurship programme. They have seen strong results because of their focus on children in underprivileged contexts and schools. Tech startups have also looked at developing AI-enabled mobile apps, including chat bots, that can support teachers in teaching literacy or teaching mathematics.
“The challenge is that there’s often a disconnect between the public education system and tech startups We need educators to be proactive in engaging with those developing tools, and we encourage students and teachers to learn how to create and design technologies that are relevant to their linguistic and cultural contexts.”
Many African governments are keen to adopt national AI strategies and integrate AI into their national policies on technologies in education. In Côte d’Ivoire, where AI is already being widely used in the private sector. Mariatou Koné, the Minister of Education, says that the country’s education system is undergoing a transformation, following a 2022 review which recommended a digitalization strategy.
“We have put in place initiatives to ensure that everyone is aware of the issue of AI. It can provide individual learning programmes, and help struggling students to improve,” said Ms. Koné. “However, we are worried about potential abuses. We have to be able to protect personal data and ensure that learners are aware of the potential dangers.”
The Minister agrees that, in order to guard against bias, the pool of engineers building AI tools needs to be expanded. “We need the right tools, adapted to the African context, to the Ivorian context. We have our own history, our own heritage. If we create our own industry, it has to be adapted to the realities of Côte d’Ivoire.”