Ask not what your government can do for you, but what AI can do for your government – The Mail & Guardian

Date:


Graphic Gga Africaai Website 1000px

(Graphic: John McCann/M&G)

If you Google “What can artificial intelligence (AI) do for government?” you will get a string of applications, including traffic flow analysis, streamlining healthcare services, enhanced large-scale infrastructure monitoring, processing large amounts of data, task automation through chatbots, cyber-attack prevention and smart policymaking.

In other words, AI can revolutionise government functions at every level, making them far more efficient and effective.

“In competent hands, and exploiting other digital technologies like blockchain, AI could create massive governance wins, especially in the arena of resource allocation,” says Dr Ross Harvey, director of research at Good Governance Africa (GGA). 

“For instance, monitoring the performance of a sewerage plant or other critical infrastructure such as stormwater drains through remote sensors could help managers implement an automated process of timeous maintenance, curtailing unnecessary expenditure.” 

The lure of AI as a means to accelerate the attainment of the sustainable development goals (SDGs) in Africa is irresistible. For example, AI and automated drones have been proven to increase agricultural productivity, directly addressing sustainable development goal 2 on zero hunger. 

Addressing goal 3 on good health and well-being, AI is set to be at the forefront of diagnostics, able to detect diseases such as cancer earlier and more accurately. And looking at goal  16 on peace, justice and strong institutions, AI can help monitor government activities, detect corruption and improve transparency by analysing large volumes of data related to public services, budgets and contracts. 

But the fuel of optimal AI performance is data — high-quality data to avoid bias and error — and this is where African countries are sorely handicapped, especially in sub-Saharan Africa, known as a “data desert”. 

Without localised data, the AI models generalise based on patterns that are irrelevant or incorrect in the African context. Healthcare AI models trained on Western data may not accurately predict diseases or conditions in Africa, and in the agriculture sector, without local datasets on soil types, climate conditions and pest behaviour, AI models will provide inaccurate predictions and advice.

The World Bank’s statistical performance indicators — data use, data services, data products, data sources, and data infrastructure from 0 to 100 — show that South Africa performs best in sub-Saharan Africa (at 82.37 in 2023), followed by Mauritius at 78.35 in the same year. Perhaps predictably, South Sudan scores worst (at 27.51). Gabon has improved the most over the past decade. But, even with good data at the national level, there is a dearth of sub-national data. 

How can these data deserts be “planted”? NGOs go some way to building up localised data resources in the governance space, for example GGA’s Governance Performance Index, which assesses the state of local governance in South Africa. 

If AI is to work optimally it requires information sharing, that is data, from citizens that can be captured. This hangs on internet connectivity. 

There are initiatives to address this such as the African Development Bank’s Desert to Power project, which harnesses solar energy in the Sahel region to power connectivity solutions, facilitating data access in remote areas.

Robust data infrastructure, including data centres and cloud services, is also needed for storing and processing large volumes of information, complemented by open data, which facilitates AI’s access to information. 

“We really need data that speaks to Africa itself, and the case for open data means we are empowering citizens and at the same time encouraging innovation and efficiency, and not using data that is inaccurate,” said Baratang Miya, chief executive of Girlhype Coders Academy, during a panel discussion titled “Fostering prosperity through policies on artificial intelligence in Africa” at the Economic Commission for Africa Conference of African Ministers of Finance, Planning and Economic Development (COM) last March.

There are risks — open data is vulnerable to misuse and exploitation for nefarious ends — but for organisations working in this space, such as the Open Data Charter, a global initiative promoting open data to improve governance and transparency, and Masakhane, a pan-African research community focused on natural language processing for African languages, the benefits of getting the continent up to speed in the fourth industrial revolution outweigh those risks. 

Sound policy, ensuring that AI is used in the public interest, will mitigate those risks, at least in the governance sector. 

“In the effervescent realm of AI, the known, the unknown and the unknowable is best addressed through governance with humanity at its centre,” noted Fayaz King, deputy executive director of Field Results and Innovation for the United Nations Children’s Fund (at the COM conference).

South Africa, Nigeria, Kenya, Rwanda and Tunisia are among the countries leading the charge in developing national AI strategies, but coordinating a unified approach among 54 countries on the continent is a complex matter, one that’s necessarily being tackled by the African Union with its new Continental AI Strategy, announced in August 2024 and to be implemented over five years from 2025 to 2030. 

The first phase focuses on setting up governance frameworks, developing national AI strategies and mobilising resources, and the second phase (from 2027) looks to carry out key AI projects and initiatives.

But, without skilled people to oversee AI and other hi-tech applications, the second phase will probably run into difficulties. 

The sea-change in the job market is already upon us: a 2020 report from the World Economic Forum predicted that AI and robotics would displace 85 million jobs by this year (2025), at the same time creating 97  million new jobs areas such as AI development, data science and human-AI collaboration.

There are unique problems in this job-shifting landscape. Automation can exacerbate inequality if the benefits primarily go to a small segment of the population rather than those losing jobs and incomes. As Daron Acemoglu and Pascual Restrepo remark in a paper titled Artificial Intelligence, Automation, and Work: “If we do not find a way of creating shared prosperity from the productivity gains generated by new technologies, there is a danger that the political reaction to these technologies may slow down or even completely stop their adoption and development.”

That said, Africa’s tertiary education institutions have been getting ahead of the game with data science and AI courses and programmes, understanding too that emerging markets have the most to gain from AI. African educational institutions, such as the University of Pretoria in South Africa, Makerere University in Uganda, and Kwame Nkrumah University of Science and Technology in Ghana, have developed prominent AI labs and published research focused on leveraging AI for social good.

“We are entering a world with job descriptions you haven’t yet heard of, like blockchain consultant,” says Miné de Klerk, leader of the design and future learning team at Eduvos, a private higher education institution based in Midrand, Johannesburg. “Our students rightly demand more future-oriented learning. To meet this demand, we must not only adopt new technological solutions but also continually evolve our approach to curriculum design.” 

In the governance space, a job title might be AI ethics and governance officer. This person would work closely with AI developers and data scientists and need to be well-versed in AI technologies and algorithms, alongside expertise in regulatory compliance and risk management. 

In tandem with educational institutions adapting their offerings, the technology ecosystem in Africa is growing rapidly, supported by investments in AI development by global tech giants such as Google, Microsoft, and IBM, and entrepreneurship development fostered by innovation hubs like CcHub in Nigeria and iHub in Kenya. 

As these developments and initiatives push the AI revolution forward in Africa, its governments will be increasingly compelled to follow suit, and the empowerment of citizens through improved governance will hopefully be exponential. Ultimately, Africa might be well placed to become a leader in using AI to solve not only local but global problems.

Helen Grange is a writer and sub-editor at Good Governance Africa.





Source link

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Share post:

Popular

More like this
Related

Gen Z needs to get out of its comfort zone

The left has become unrecognisable, not least because...

Netanyahu gives Trump ‘golden pager’ in apparent reference to Lebanon attack – report | Donald Trump

Benjamin Netanyahu reportedly gave Donald Trump a “golden...