Fight African tender corruption with Artificial Intelligence!

“Show me a tender, I’ll show you corruption,” a witty phrase to an uncanny problem common across the African continent.


Throughout the continent’s developmental agenda, contracted work has been the backbone of vast infrastructural development, public goods procurement and service provision to governments. It has also been the source of vast graft that robs the poor and enriches the few politically connected and economically well off.
The origins of the tender system to African governments is somewhat inconclusive. It seems such that post-colonial development, which was largely funded and encouraged by former colonial masters through their own developmental agencies, created a system of consultants for just about anything to do with government administration. These consultants were usually paid by these governments to advise on projects funded by those agencies, but in reality most of the developments benefited the same colonial powers. As the continent developed, the system became entrenched in these newly independent states along with the culture of corruption which is now endemic in today’s Africa .
Today, governments all over the continent issue tenders for everything, from milk supply to building power stations. Whilst the system was designed to create a level playing field for the development of local capacities, foreigner companies still play a major position in the contract space. The system would normally dictate for a committee of civil servants to evaluate and make recommendations to award said tender. The human element in the whole system then becomes the cog. Human manipulation, violation and gross influence of the system breeds what the continent has come to be famous for, tender fraud and corruption.
According to Transparency International, Africa remains the leading continent in terms of corruption. Whilst Botswana does relatively well, out-ranking the rest of the continent and being globally competitive, the rising number of cases brought forward to the courts of law by the Directorate on Corruption and Economic Crime (DCEC) show that we are increasingly catching our neighbours’ virus.
Enter, Artificial Intelligence, A.I a robust set of computer programs designed to predict outcomes based on big data analysis. In the case of tender fraud and corruption, the biggest problem usually lies in the human element, A.I takes care of that as it removes the need for a tender committee decision.
According to global think tank K2 Intelligence, “A.I is based and built on learned behaviour, as opposed to instinctive behaviour which is unique to humans.” This means that it is programed to behave and react in a particular way, however with a lot more discretion based on set parameters. It involves big data, digital forensics and other integrated tools for such a system to be successfully implemented.”
A new programme by University of Valladolid researchers in Spain, a Western nation grappling with its own corruption scandals, was put in place to predict the propensity of graft at local government and municipal levels. The research was published on the Social Indicators Research journal under the title, “Predicting Public Corruption with Neural Networks: An Analysis of Spanish Provinces”.
The breaking ground study by the researchers indicated that for corruption to be adequately eradicated, there is need for early warning systems, similar to those in any disaster prediction scenario.
“We contend that corruption must be detected as soon as possible so that corrective and preventive measures may be taken. Thus, we develop an early warning system based on a neural network approach, specifically self-organizing maps, to predict public corruption based on economic and political factors. Our model also provides different time frameworks to predict corruption up to 3 years before cases are detected.”
It is such a system that would help Africa’s plight of gross graft by providing the necessary tools to actually act before corruption take root.
The Bloomberg Law Business outlines it a lot more explicitly, noting that A.I would enable corruption busters to spot “suspicious digital activity, anomalies in accounting entries and transactions, and uncovering relationships among potential parties to a fraud that they may not have been able to discern in the past.” It is through such tools, graft trails would become less complicated.
Adopting any form of A.I to government procurement would mean having to change present legislation. For Botswana, chief amongst would be the Public Procurement and Asset
Disposal Board (PPADB) Cap 42:08] of 2001 act.
Whilst we can never discount threats to A.I such as data manipulation and hacking, the combined efforts of vigilant human intelligence and A.I would improve the fairness of tender awarding and management, possibly extending to all elements of government procurement and financing.
It will take political and bureaucratic will for such a solution to work on the continent, whilst pessimism would probably be the first reaction, but the possibility of success to could truly transform the development agenda on the continent.