Select a location

This selection will switch the site from presenting information primarily about Senegal to information primarily about . If you would like to switch back, you may use location selection options at the top of the page.


Aiscension: AI in the legal sector

Africa Connected: Issue 6

By Werner Rysbergen

The rise of technological developments has brought with it increased, and new forms of, risk. In today’s digital world, we are creating more data year on year. Data storage, ease of communications (including the use of email and chat), along with the ubiquity of photo and video sharing means that data volumes are increasing exponentially. New technologies pose new challenges and require new regulatory solutions. This backdrop makes it harder for businesses to monitor existing risks.

There are two concurrent and somewhat symbiotic developments taking place in the legal sector that address this. On the one hand the sector “must move beyond providing cost-efficient legal advice, and focus on the speed with which it develops appropriate guidance and efficient governance so business partners can quickly make good risk-adjusted decisions and pivot with changing conditions.”1 On the other, the use of AI in legal practices is already occurring and is poised for even greater adoption. Like other technological advances before it, AI is “evolving the practice of law and improving how in-house teams work and process information. AI can help legal departments streamline and automate tasks and processes, and in turn reduce costs and increase productivity.”2

One area of risk which AI can help mitigate, and detect, is cartel activity. DLA Piper has recently launched Aiscension to help detect and prevent cartel activity in collaboration with Reveal Data. Before exploring Aiscension further, let’s first take a closer look at the problem it solves.

Cartel enforcement in Africa

A cartel is an agreement between competitors to rig the market in some way. The most common forms of cartel conduct result in price fixing, market allocation, bid rigging and collective boycotts. Cartels reduce firms’ incentive to innovate and / or to price competitively. Consequently, they cause consumer harm, and are considered to constitute the most serious form of competition law non-compliance.

Competition law enforcement in Africa has developed rapidly over the past few years. We continue to see the establishment of new domestic and regional competition authorities, as well as existing competition authorities introducing amendments to competition legislation strengthening their enforcement powers. The focus of newly established competition authorities is typically to build capacity and experience through merger control work. The focus of newly established competition authorities is typically to build capacity and experience through merger control work, and over time the focus has then shifted to the enforcement of restrictive practices such as cartels.

The Competition Commission in South Africa has achieved great success in combating cartels. A significant part of its success is due to its corporate leniency policy offering whistleblowers immunity (or a significant discount ) from fines if they cooperate with the Commission and provide credible information to prosecute the remaining cartelists. Administrative penalties of more than ZAR300 million were imposed by the South African competition authorities for each of the past couple of years, while in 2016/2017 penalties of more than ZAR1.6 billion were imposed. Aside from the financial implications of being found guilty of cartel conduct, the reputational harm caused to firms is severe and long-lasting.

South Africa is certainly not the only African country focused on prosecuting cartel behavior. Investigations into cartel behavior (including the carrying-out of dawn raids) and the prosecution of firms have taken place in many other countries, including Botswana, Egypt, Kenya, Mauritius, Malawi, Namibia, Tanzania and Zambia. Industries that have been subject to investigations include insurance, healthcare, construction and oil and gas. Many of the African competition authorities have existing corporate leniency programs or introduce them once they have identified a specific sector or industry they intend to investigate. Administrative penalties of up to 10% of turnover can be imposed, and criminal sanctions are available in a number of jurisdictions (for example in Botswana, South Africa and Zambia).

Aiscension for cartels

Cartels are also, by their very nature, secretive and hard to detect. However, the introduction of AI to this space changes that. By using the power of Reveal Data’s first class neural-net AI, along with the data and know-how available within a global law firm like DLA Piper, the AI has been taught to spot these cartel risks and enable our lawyers to quickly run a review and advise clients of their cartel risks. Specifically, Aiscension has been trained to uncover the following forms of cartel behavior: price fixing; bid rigging; market sharing; collective boycotts and exchanging competitively sensitive information.

Aiscension provides the ability to detect risks and helps to:

  • identify cartel risks within businesses before they become regulatory investigations;
  • maximize the chances of securing regulator leniency or immunity; and
  • minimize disruption to core operations in a cost-effective manner.

Human versus machine

“AI and cognitive computing or machine learning are generally interchangeable terms that all refer to how computers learn from data and adapt with experience to perform tasks. In-house attorneys’ anxiety over AI often stems from concerns that it will replace them or the work they do. Where the evolution of AI can play a significant role in the legal industry is by augmenting lawyers’ work and help increase their productivity – not replace them.”3 Aiscension is the perfect example of this in action. The combination of AI and DLA Piper’s experience sets Aiscension apart from other solutions. Aiscension is a closed-loop system that harnesses proprietary lawyer-trained AI models to detect risk factors in vast sets of data – with lawyers assessing the results to advise companies on how to proceed. This represents the best of both worlds.

By Ilan Sherr (Legal Director, DLA Piper UK LLP), and Werner Rysbergen (Director, DLA Piper Advisory Services (Pty) Ltd).

For more information on cartel trends across Africa and Aiscension, please contact Werner Rysbergen and Ilan Sherr.