North American researcher Nathaniel Allen told Lusa that “There are several forms of Africanizationr” a Artificial Intelligence, a technology that, even though it is being developed by the greatest world powers, constitutes “a great opportunity” for African countries.
"It's true that companies and individuals in the United States and China are at the forefront of developing artificial intelligence (AI), especially for military and security purposes, but those who create the technology are not necessarily those who reap the greatest benefits.“, the researcher and professor of Security Studies at the Africa Center for Strategic Studies (ACSS), in Washington, told Lusa.
Nate Allen acknowledged that, in many ways, many of the AI algorithms and the way the technology is being used “have the potential to – and, in some ways, are already – reinforcing inequalities between the North and South of the world. ”, but the development of digital technology “is not like building a warship or an airplane, which depends on huge economies of scale. Digital technology is cheap and can spread very quickly.”
Even though it is “cheap”, as Allen says, its operation “is extremely expensive”, says Abdul-Hakeem Ajijola, president of the African Union (AU) Group of Experts who designed the Malabo Convention on data protection and ethical use of AI on the African continent.
"I think ChatGPT — Open AI's platform — costs around 100.000 US dollars per day to run, which implies that having an AI platform is not something that most African organizations and even African governments can afford“, the Nigerian expert told Lusa.
Artificial intelligence is ripe to help solve some important problems in Africa, from agriculture to the health sector, including the financial system, but Senegalese expert Seydina Moussa Ndiaye warned of a new “colonization” of the continent by this new technology, if large foreign multinationals continue to feed on African data without involving local actors.
For Ndiaye — who participated in the development of the AU's Pan-African AI Strategy, as well as a consultant for the United Nations Global Partnership on Artificial Intelligence (GPAI) —, this is indeed the “biggest threat” of artificial intelligence on the African continent.
BUT: Artificial Intelligence can help improve privacy, guarantees Government
“For me, the biggest threat is colonization. We may end up having large multinationals in the field of AI that will impose their solutions across the continent, leaving no room for the creation of local solutions,” he told UN News in a recent interview.
Most of the data currently generated in Africa is owned by multinationals whose infrastructures are developed outside the continent, where most African AI experts also operate.
“This leads us to think that we are facing a new wave of neocolonialism, because, let's be realistic, most algorithms are developed in Southern California or in the United States, in general,” Ajijola also said.
Another important element to take into account is the context of the fourth industrial revolution, warned Ndiaye. The power of AI, combined with advances in biotechnology or technology, is immense, but all solutions need to be tested and Africa could be where companies look.
“If left unsupervised, we could end up having tests on humans with chips or even integrated biotechnological elements that we have improved. These are technologies that we have not mastered well. In regulatory terms, there are certain aspects that were not considered. The framework for applying existing ideas and regulations is not effective”, according to the Senegalese expert.
“In concrete terms, and when these things are not controlled, it can happen without anyone knowing. We could have Africa being used as a guinea pig to test new solutions, and this could be a great threat to the continent”, said Ndiaye.
The data sets, the specialized knowledge, the training of the algorithms are, for the most part, based in the United States and China, “which is certainly a danger”, Allen conceded.
While he “understands” Ndiaye’s concerns, Allen said he believes that “once certain prejudices are identified, it is possible to mitigate them.”
“I'm not sure if the narrative that says that the world's great powers will be the only ones to benefit from the dissemination of technology, even if they develop it, is completely accurate,” he said.
