Kathy Gibson reports from the AI Everything Summit in Nairobi – Artificial intelligence (AI) governance and policies are top of the agenda for African countries as they seek to grasp the many opportunities that the technology promises.
There’s no shortage of advice, ideas and precedents – but knowing which are relevant and which won’t work in the African environment can be a challenge.
To help African policymakers, the International Telecommunications Union (ITU) offers its ITU AI for Good Sandbox.
This provides capacity building, AI research, and experimentation for standards, policy and strategies, self-assessment toolkits, and onboarding resources for countries.
In addition to sharing experiences on collaborative work towards common understanding of vocabulary, common terminology of AI Readiness and the importance of interoperable AI solutions using standards.
Thomas Basikolo, machine learning program coordinator at the ITU, Switzerland, explains that these tools are part of the ITU’s attempt to work with countries to bridge the digital divide.
Governance is key to the successful implementation of an AI strategy, says Professor Lourinho Chemane, chairman of INTIC, Mozambique.
The country has set up a national AI commission to drive governance and is currently drafting an AI strategy.
“Everyone agrees on the positive impacts of AI, but also on the risks and challenges,” he says.
Chief among these is the need for inclusivity – particularly in the matter of local languages – and Mozambique is looking to identify policies that address inclusion.
Capacity building is another key element, and the country aims to ensure that skills development addresses all levels of competency – from basic users to high-level developers.
Domain experts are also being drafted in so AI capacity can be built in vertical markets like healthcare, agriculture and justice.
“It is important that we facilitate the inclusion of AI in people’s day-to-day activities,” Prof Chemane says.
Dr Nkundwe Mwasaga, director general, ICT commission in the Ministry of Communication and Information Technology, Tanzania, agrees that skills are important. However, he adds that digital security and trust are also vital.
“If all of these things are done well, the transformation will reach its objective,” he says.
Also important is the need to have all the stakeholders at the same table, Dr Mwasaga says.
He recommends that any AI policy ask these questions:
- What is the problem that AI is the solution to?
- Whose problem is it?
- What new problems will be created?
- What people and which institutions will be most harmed?
- What is being gained or lost cultural-wise?
- What shift in economic and political power will take place?
- What intended and unintended use cases will there be?
But African countries don’t have to reinvent the wheel. They can draw on experiences from other countries that have walked the road and discovered where the pitfalls are.
Jincheng Li, China, says AI belongs to all people and China is ready to share its experiences and solutions with other countries – especially those in the global south.
“We are willing to walk with other countries on joint technology research,” he adds. “We need to all work together to ensure that AI serves the common good.”