AI and Mobile Money: Bridging the Financial Inclusion Gap

In our recent blog, we defined artificial intelligence (AI) as the ability of a machine or computer to emulate human tasks through learning and automation. This blog discussed key opportunities for AI in the attainment of the UN Sustainable Development Goals (SDGs) while giving practical examples of how this has been effected to date. We then published a blog highlighting the use of frontier technologies in developing countries in the field of digital health.

Our latest blog in the series focuses on how AI can be used to advance the use, and value of, mobile money services in helping bridge the financial inclusion gap.

In 2019, the mobile money industry surpassed a billion registered accounts and recorded close to $2 billion in daily transactions. The industry provides financial services to the previously underserved and addresses key developmental challenges in most of its markets. As the mobile money industry continues to grow and specifically seeks to leverage new technologies, we recognise that there are areas where AI can help bridge the financial inclusion gap.

How can AI be used in mobile money?

AI often serves as a catch-all term for a wide-ranging set of technologies, including machine learning, big data analytics, statistical modelling, robotics process automation, natural language processing, and speech or object recognition. It typically combines machine learning processes with big data analytics to operationalise new methods of synthesis and analysis of broad datasets.

The 2020 World Economic Forum Global AI in Financial Services Survey identified five key areas where AI can contribute to the growth of financial services, which are also relevant to the mobile money industry.

These are as follows:

  • Generation of new revenue potential – The use of AI-enabled data analytics, has been effective in generating novel insights for competitive advantage and product differentiation. Organisations like Branch[1] are using AI algorithms and alternative data sources for credit scoring which creates new products and services outside of the typical lending facilities. This enables providers to offer a wider range of financial services to the previously unbanked.
  • Risk management – AI can increase analytical capabilities in risk management and compliance to improve speed and lower costs of detecting fraud and identifying potential risk factors. MyBucks, a leading FinTech company, is now deploying an AI fraud detection system called Dexter that is able to characterise client responses and detect irregular behaviour by monitoring various data sources in the prevention of fraud.
  • Process re-engineering and automation – AI can be used in the automation of core operational services that will lead to increased efficiency for the financial service provider. Tyme Bank is the first fully digital bank in South Africa and adopting AI in their operations has enabled them to scale rapidly, acquiring roughly 670k customers in just a few short months.
  • Customer service – AI-enabled communication channels have led to an increased use of chat bots by providers, which come close to replicating real human interaction, while generating insights for improved service delivery. The recently launched MTN chatbot for mobile money services is one example which leverages messaging and artificial intelligence to deliver customer engagement and enhance users’ MTN MoMo experience.
  • Customer acquisition – AI has various uses in customer acquisition, including making outreach more personalised and speeding up on-boarding procedures. Safaricom’s chabot, Zuri, helps users manage their marketing messages according to preference. This allows Safaricom to provide relevant products and services for previously unbanked users.
Considerations while adopting AI

As with any new technology, there are transitional challenges that need to be addressed in order to see the benefits of AI in the mobile money industry.

We highlight some of them below:

Regulatory uncertainty
The financial services industry has long been subject to complex regulatory requirements. Adopting AI ethically will require regulators to engage closely with mobile money providers to increase transparency and ensure compliance with the relevant legal and regulatory obligations. In our recent report titled Demystifying regulatory concerns for the use of cloud services in mobile money, we explore some of these regulatory concerns and identify practical steps for regulators and mobile money providers to address concerns around the use of the cloud, and ultimately, artificial intelligence.

Skills gap
To seize the benefits of AI, such as productivity and efficiency in service delivery, there is need for a new wave of professionals to do the work. In a recent EY study, 31 per cent of the respondents believed that lack of skilled personnel would be the greatest barrier to AI implementation. Mobile money providers (MMPs) and other key players in the sector will need to think innovatively to address this concern today. Partnering with entities that are already providing AI services could address the immediate skills gap.

For a longer term and more sustainable approach, MMPs could consider engaging with partners who would be able to deliver trainings and build capacity in this area to reskill and upskill staff. Having a rightly skilled workforce will also ensure that the use of AI does not inadvertently impede financial inclusion initiatives. In some instances, a lack of understanding of AI and reliance on low-quality data sets can lead to biases and discriminatory practices, as was discovered by Amazon.

Organisational change management
In addition to bridging the skills gap and addressing regulatory uncertainties, mobile money providers will need to couple AI deployments with adequate change management programmes in order to ensure that investments are not lost. A CIO article, The human side of implementing AI, discovered that only 20 per cent of businesses use AI at scale. This is largely due to misperceptions or a lack of understanding held by the employees of such organisations. In order to fully maximise the benefits of AI, mobile money providers would need to adopt a people, process and technology approach to change management, and involve employees as early as possible.

Going forward

AI has the potential to expand access to mobile money services improving access to financial services, this will inevitably lead to socio-economic benefits for society and the global economy. Currently, a lack of regulatory clarity and skill shortage are key areas that players in the mobile money industry will need to address. Additionally, the ability to effectively and strategically integrate AI into business processes will need careful change management strategies so as to derive maximum benefits for companies and consumers. Going forward, we hope to engage deeper with the industry to find practical solutions to some of these challenges for the mobile money industry.

[1] Mobile money providers recognise the risks associated with the use of digital credit services, and are constantly working to improve their operations and guide their partners to mitigate existing and emerging risks, for consumer trust. GSMA (2019). Digital credit for mobile money providers: A guide to addressing the risks associated with digital credit services

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.