Artificial intelligence, or machine learning (the ability of a computer to learn without being programmed), enables companies and individuals to make calculated, strategic business and personal decisions through the manipulation of enormous volumes of data. The predictive analysis tools built into the technology allow for truly intelligent future forecasting.
This means that millions of historic purchases in a retail environment, customer behaviour in the insurance sector or simply getting home from work can be better analysed and manipulated to produce meaningful reports that offer usable likely future scenarios.
The idea of machine learning seems more complex than what is should be – ordinary consumers are already using AI without even realising it.
Apps such as Siri (and similar voice activated apps), Google Maps, the popular photo editor Prisma and the various language translation tools already available are just a few examples of technologies emerging to make people’s lives a lot easier (or more interesting in the case of Prisma).
A new app called Pounce allows users to photograph people wearing certain clothing and provides them with links to sites selling similar items (or even the exact item) online.
Soon, controlling appliances at home from a remote location (e.g. televisions, lights, curtains) will be commonplace. We will see far lower death rates from accidents as a result of self-driving (autonomous) cars.
Almost any business process can be improved by the analysis of data and learning of patterns done by AI. For instance, banking customers may experience less fraud as the advanced tools pick up on suspicious transactions with a high level of precision and doctors will be able to improve the detection and treatment of diseases like cancer.
Still in their infancy, these tools need to be taken through their paces and fine-tuned in the corporate world.
Some industries, such as financial services, healthcare, retail and telecoms already make use of the technology to improve their ability to produce relevant products or services, and ensure that customers are served better, but to varying degrees of success.
Credit providers and venture capitalists are starting to analyse entrepreneurs who might not have a credit record and provide them with access to opportunities to growing their businesses based on real, detailed insights.
Some sectors in South Africa are doing better than others in this regard, such as the financial services industry, while others such as the insurance industry are lagging behind.
This is because AI is an emerging skill-set that is not easy to come to grips with. There is likely to be a severe shortage of these highly specialised skills in the future, especially with massively increased demand. Globally, big corporates are acquiring smaller companies with specialist skills, and using those to train internal team members.
In theory AI infrastructure should reduce incumbent IT infrastructure, meaning more efficient use of hardware to analyse data. However, the analysis of huge volumes of data (e.g. transactional data) is costly. Since most cloud based cloud hosting is internationally based, and charged for in dollars, it can be expensive for South African companies.
Still, AI is becoming more accessible through the proliferation of open-source tools and vendor-provided AI solutions tailored to specific business problems.
In the medium to long-term it will become more pervasive across the world economy, subtly improving many existing processes and enabling completely new services.