Africa needs to think strategically about what AI and innovation means for the continent, encourage capacity building, and provide adequate infrastructure and data collection avenues.

This is according to the State of AI in Africa report 2023, prepared by the Centre for Intellectual Property and Information Technology Law at Strathmore University in Kenya.

The report says Africa is embracing AI in ways unique to the continent and is showing no signs of slowing down.

It notes that AI development is occurring in Africa and the highest rate of development of AI-enabled tools is found in the agriculture, health, financial services and education sectors.

Africa has more than 2,400 AI organisations operating across various industries, including health, wellness and fitness, agriculture, legal and judicial services, skills development and insurance, it says.

The report highlights the potential of AI technologies to transform these and other sectors in Africa, such as business operations, education, and transportation.

It focuses on four key thematic areas: AI and data; AI and innovation; AI use and impact in health, agriculture, legal and creative fields; and responsible use of AI.

The report also highlights the challenges faced in AI adoption in Africa, including ethics, skills and capacity, ill-equipped policy frameworks, and a need for a structured data ecosystem.

It also cautions that the absence of robust policy frameworks leave AI deployment largely unregulated, and ethical concerns, such as accountability, data bias, transparency, and socio-economic implications, arise.

To this end, the report recommends that to ‘fully realise the potential of AI and innovation in Africa, countries may need to work together to establish the necessary infrastructure.’

It says: “There is a need for more comprehensive AI policies and frameworks to promote responsible innovation, address ethical concerns, and ensure equitable access to AI benefits across all communities.

“African governments need to put in place privacy policies that safeguard data subjects against the adverse effects of data breaches in the utilisation of AI tools and products. There is a need for clear policies and guidelines on limits of surveillance in AI.

“Africa needs to think strategically about what AI and innovation means for the continent and encourage capacity building, and provide adequate infrastructure and data collection avenues.

“For this move to be successful, then it must stop copying and pasting techniques from the Global North and focus on its unique challenges and solutions in AI and innovation fields.”

Governments, the report says, need to support the development of a responsible AI ecosystem in the following ways: “being willing to adopt AI, and being able to adapt and innovate to do so; promoting a good supply of AI tools from the technology sector; and ensuring these tools are built and trained on high-quality and representative data, and building the appropriate infrastructure to be delivered to and used by citizens.”

In addition, it says, national policies and regulatory frameworks need to ensure that emerging technologies benefit humanity.

The report states: “Responsible AI principles must ensure that fairness, transparency, and accountability are integrated as the underlying principles in the development, adoption, implementation, and use in consideration of human-centred AI.

“There is a need for African developers to develop ‘African AI’ that would be tailored to the contextual needs of the continent. African states should look towards creating a conducive environment for the start-ups to thrive.”