In February, simply earlier than the AU’s AI coverage draft got here out, Shikoh Gitau, a pc scientist who began the Nairobi-based AI analysis lab Qubit Hub, revealed a paper arguing that Africa ought to prioritize the event of an AI business earlier than making an attempt to manage the expertise. 

“If we begin by regulating, we’re not going to determine the improvements and alternatives that exist for Africa,” says David Lemayian, a software program engineer and one of many paper’s co-authors.  

Okolo, who consulted on the AU-AI draft coverage, disagrees. Africa ought to be proactive in growing rules, Okolo says. She suggests African international locations reform present legal guidelines reminiscent of insurance policies on knowledge privateness and digital governance to deal with AI. 

However Gitau is worried {that a} hasty method to regulating AI might hinder adoption of the expertise. And he or she says it’s important to construct homegrown AI with purposes tailor-made for Africans to harness the ability of AI to enhance financial development. 

“Earlier than we put rules [in place], we have to do the exhausting work of understanding the complete spectrum of the expertise and put money into constructing the African AI ecosystem,” she says.

Greater than 50 international locations and the EU have AI methods in place, and greater than 700 AI coverage initiatives have been carried out since 2017, in accordance with the Organisation for Financial Co-operation and Improvement’s AI Coverage Observatory. However solely 5 of these initiatives are from Africa and not one of the OECD’s 38 member international locations are African.

Africa’s voices and views have largely been absent from world discussions on AI governance and regulation, says Melody Musoni, a coverage and digital governance skilled at ECDPM, an independent-policy suppose tank in Brussels.   

“We should contribute our views and personal our regulatory frameworks,” says Musoni. “We need to be customary makers, not customary takers.”