Byte Sized Tech #1: How AI is Transforming Businesses Across Africa

Listen to the full podcast

In your expert opinion how is AI transforming businesses across

AI is basically having a profound impact on how it transforms businesses of all sizes, so we are seeing AI automate tasks, increase efficiency, and gain deeper insights in their customer experiences. What we are seeing out of that is that organizations can focus more on strategic initiatives instead of focusing on the more mundane day to day tasks that they had to do before

Can you share some specific examples?

Absolutely, so within the African landscape we see AI transform some key economy drivers, like with agriculture you see driving precision farming through accurately predicting weather patterns and optimizing irrigations models. And in finance we see AI quicker detect fraud and create a more efficient loan application processing. And in medicine we see AI detect medical images to drive quicker diagnostics. So we can say the list is limitless,  one within our space like cybersecurity we see better threat detection and remediation. So as long as you get data to drive decision making you can utilize AI in any industry

There is a very burning question in AI, which is around the challenges organizations face in adopting this technology could you shed some light on that

Yes, of course every opportunity has its challenges, despite AI’s fast potential the adoption and implementation in Africa faces several challenges, so for instance we are seeing lack of relevant skill, inadequate basic and digital infrastructure, lack of sufficient investment, and less dynamic regulatory framework around things like ethical standards.

So for the skills gap, we see a lack of relevant skills in the population which is a major barrier to adoption, because it means that we cannot develop AI driven solutions, and I see there the opportunity where partners like Incentro who already have experience and expertise can contribute develop and growing talent, which can then help develop these new AI solutions

Data limitation there is a lack of high quality and diverse data, and it is important for AI to gain this relevant data of the population so it can make solutions which are tailored for set population. So we need to clean up data to the level where it is mature to grow AI models. And to be honest we still don't have enough investment yet in R&D.

We also see a lot of fear of displacement and replacement of jobs, which is an idea we have to get rid of, as we should see AI as a tool to do more important and meaningful tasks ourselves whilst leaving the rest to AI. so looking ahead I can say AI presents a good opportunity for Africa to leapfrog. And it will allow us to drive our key SDG goals, in agriculture, finance, health, and education

Would you say we are apprehensive of this new technology

I’d say it's rather a lack of knowledge, it's going to transform the way we work either way, and its for the good. So if we train ourselves to see the value, it will change the mindset. I also think the skill gap is a contributor, because I don't know how to use it, I also have fear of creating time to do that and how it will impact my input

Who is responsible for overcoming this fear, is the government, the private sector, the providers of this tech. Like if I like a show on netflix I am actually training their AI model, whilst not everyone might be aware that that is what’s happening behind the scenes, so should we shed a light on that AI is already part of our lives?

I think everyone has a stake. It's important that people will start to learn this in school, it's important for governments to have the right ethical frameworks in place, and for partners like Incentro have the role to build capacity, to insure that people have the right skills to adopt AI. And ofcourse now within organizations it's important to have AI as a part of your day to day operations and businesses, and identify which areas of your business can be automated or improved with AI, so that becomes part of our day to day life and not something new or something that comes on the side that he have to learn and adopt.

One last question, in your view, which industry stands to benefit the most from adopting AI?

I’d say all industries to be honest. Ai is becoming the core of every business, because as long as you have a process A to B as a business, AI can be part of that. 

So as I recommended earlier, the most important part is every industry identifying which elements and pain points can be transformed by AI and how that can be applied. Most customers we work with basically are already seeing the value of different use cases, some are looking to automate the day to day like writing emails and reports, whilst others want to use AI as chatbots to improve their customer experience. 

So I'd say any industry can basically leverage AI to transform how they work.  And of course, the more data that you have, the more data that you gather,  the more accurate your model will be.

But let me ask you this one last time. It can be a challenge to sort of structure existing data, correct?  Especially if you have a vast amount of data spanning from years and years and years. Now, as for example, a financial institution, would it make sense for me to start collecting data today? Or should I leverage the existing data, which would be more beneficial in general? To either put the resources towards developing a structure model that you can train with the existing data or start from scratch and develop a model that is unbiased and can be utilized by decision makers. 

Yeah, so I'd say both can work. And in that sense, I'd recommend working with also your legacy data. I'd say, but that would of course require support and maybe a little bit of experience from someone who has done this before. So what that would mean is basically you collect both your structured and unstructured data, and then you put them in a form that can be used to analyze.

And of course, now later on develop the AI models. But as I said, I think that's still the biggest challenge we have in this region where data is still sitting in silos. Data is not regulated.

We have not very clear data regulation policies. So people are even afraid to explore at that very level. And AI cannot be successful if you don't have clean data. 

So it means we should start at that level where we have the right regulatory framework when it comes to data. We work collaboratively to create clean data rooms. And then from there, we can build things that are basic as analytics, right?And those are the things that will drive better decision making for us.

That will also make a difference in organizations to eventually see the value of AI

I completely agree. I know the costs of time and training all of this can be quite daunting for many, many decision makers, especially those in their later stages of life.

I feel the regulations need to be picked up. And skill sets need to be further developed. And Incentro, of course, as well, can go ahead and start leveraging some of these tools that are provided to them by Google, for example, to sort of help existing adopters of AI to really, really leverage and spread the word that this is truly a revolution that's taking place.

Thank you, Liz.