Artificial intelligence projects now prolific across SA firms

Read time 5min 10sec
Local companies expect to gain significant financial benefits by using AI solutions to optimise operations.
Local companies expect to gain significant financial benefits by using AI solutions to optimise operations.

During the past decade, South African organisations invested $1.6 billion (R23.5 billion) in artificial intelligence (AI) technologies.

This is according to the AI Maturity Report in South Africa, commissioned by Microsoft and conducted by Ernst & Young. The findings were revealed yesterday at Microsoft SA’s offices in Johannesburg.

The new research shows AI pilots and experimentation are now prolific across South African companies, with businesses showing a willingness to embrace AI and experiment using new technology.

In fact, 46% of South African companies say they are already actively piloting AI within their organisations.

The study is based on surveys, interviews and case studies from 112 companies across the Middle East and Africa (MEA).

In SA, the study surveyed brands like MTN, Discovery, Standard Bank, De Beers, TymeBank, BCX, Hello Group, Nampak and Medscheme, among others.

Regional giant

Of the countries surveyed, South Africa is the third highest regional investor in AI. In the country, businesses are optimistic about AI’s potential to improve operational efficiencies and productivity, and create new revenue streams.

It trails behind Turkey ($3.4 billion) and United Arab Emirates ($2.1 billion).

While concerns still linger around creating a culture that truly embraces AI, especially given the employment sensitivities, SA’s enthusiasm to pioneer new AI solutions is the result of a very open culture towards AI, the report says.

At least 25% of South African companies consider themselves highly competent in their willingness to experiment with and learn about new technologies, compared to just 14% across the region.

Businesses are experimenting with a range of different technologies, including chatbots, robotic process automation and advanced analytics. Some 67% of South African organisations pinpointed machine learning as the AI technology most useful to them, followed by smart robotics and biometrics.

“Many AI experts argue it’s not simply a lack of technical skills that slows the progress of AI, but also a greater need for a culture of experimentation. Though AI is in its early stages of development in SA, it bodes well for AI maturity in the country that businesses are actively experimenting with exciting new AI use cases,” says Lillian Barnard, MD of Microsoft SA.

South Africa’s overall investment in AI is significant, with $1.6 billion invested over the past 10 years.

Together with the United Arab Emirates and Turkey, South Africa was one of three countries in the region with the highest AI investment activity in the last decade, the report notes.

The bulk of this investment went towards the Internet of things and social media, followed by planning, scheduling and optimisation, as well as smart mobile.

Lillian Barnard, MD of Microsoft SA.
Lillian Barnard, MD of Microsoft SA.

Financial gains

According to the study, when it comes to the future benefits of AI, companies across the MEA region are highly positive – and perhaps none more so than South African organisations.

It found that 96% of local businesses expect to gain significant financial benefits by using AI solutions to optimise their operations.

The top use cases for AI listed by companies surveyed include automation (83%) and prediction (70%). The use cases in question include a broad range of applications, from increasing employee productivity, to predicting customer churn or consumer conversion rates and proactively managing machinery downtime.

Although 42% of South African businesses say AI is one of their digital priorities, not a single company indicated it was their most important priority.

There are a number of core competencies South African companies feel they still need to address in terms of AI maturity.

While data management scored well as a competency in SA (almost a third of companies consider themselves highly competent), businesses typically rate themselves as only moderately competent when it comes to leadership capability.

Combining AI with emotional intelligence is another significant challenge for South African companies, with many respondents citing concerns around human-machine interaction, and how AI outputs should be integrated with operational processes and the people involved in those processes.

Perhaps most importantly, Microsoft says discussions around AI need to take place across all levels of South African businesses. It notes that while 83% of organisations report direct involvement at the C-suite level, this number is significantly lower (29%) at a non-managerial level.

Human resistance

Currently, 54% of companies rate impact on personnel as the top business risk in implementing AI.

While SA’s unemployment rates are high, understanding of AI is low. Microsoft explains this typically means much of the workforce’s excitement around AI is quickly replaced by fear of job losses.

A recent study by Deloitte found that despite numerous pronouncements that the fourth industrial revolution will create new jobs, South African millennials are pessimistic and uneasy about their future career prospects.

“To realise the true value of AI, organisations need to understand the scope and risks specific to them,” says Brian Lewkowicz, lead on intelligent automation at Ernst & Young Africa.

“Then they need to define the value and capabilities needed to integrate, activate and incorporate intelligent, robotic and autonomous capabilities. We hope this study opens doors for South African organisations to use AI to improve business processes and accessibility for non-technical users.”

As such, Lewkowicz adds, companies should implement change management to help employees across the organisation understand that AI helps enhance productivity and can reduce the amount of time they spend on menial tasks.

“Microsoft has always believed AI strategy should be designed with people at the centre. When companies opt for a ‘people first, technology second’ approach, human capabilities are extended and people can spend more time on creative and strategic endeavours so, ultimately, companies achieve more,” Barnard concludes.

Login with