AI models can spur enterprises


As big tech firms dominate the African digital ecosystem, we are experiencing a rise of discriminative algorithms, market dominance abuse and exploitation of workers. FILE PHOTO | SHUTTERSTOCK

The buzzword these days is artificial intelligence (AI). From Silicon Valley to Silicon Savannah and beyond, generative AI appears to be on everyone's lips.

Many people believe that generative AI is a recent development, but it has been a long time in the making.

Large language models are fuelling the current AI frenzy. These are systems like ChatGPT, which are based on deep learning algorithms that can recognise, summarise, translate, and generate text content based on vast amounts of data.

Large Language Models (LLMs) are the result of decades of incremental growth in our approach to interacting with computers.

We have progressed from using low-level programming languages to instruct machines to the graphical user interfaces (GUIs) that we use every day.

LLMs such as Chat GPT usher in a new era of human-computer interaction in which natural language as we know it is the primary mode of communication.

LLMs not only allow users to communicate with machines using text but in ways that resemble speaking to another human.

As a result, in addition to the GUI (on-screen icons), large language models have the potential to further democratise the use of AI given how much easier it is for anyone to communicate with and instruct machines.

Small and medium-sized businesses (SMBs) have numerous opportunities to improve efficiency and set themselves up for success.

Improved computing capacity can help them at every stage of their operations, from planning to execution.

One way they can benefit is by reducing the amount of time spent on repetitive tasks, allowing business owners and their teams to focus on more important tasks.

Although useful in all areas of business, some may benefit more than others. Customer service, for example, is frequently plagued by repeated questions and requests from many users.

Based on the data sets they have, LLM systems can interpret customer requests and offer guidance. A more complex or difficult to fulfil request can then be handled by a human.

In this case, artificial intelligence is not replacing humans, but rather increasing their productivity by reducing their mental load.

In the business world, there is no turning back from using large language models in computing. Micro, small, and medium-sized businesses stand to benefit the most from these developments.

LLMs have the potential to assist SMBs in staying competitive with much larger enterprises with greater resources.

Furthermore, these LLMs can enable business owners to use computer tools without necessarily possessing the high-level skills that are sometimes required for effective use.

Many people, for example, have great business ideas but lack the knowledge to write a business plan, enter the necessary financial projections, and create a pitch deck for investors.

Businesses can transform an idea into a reality with simple instructions to LLM-powered tools and a little data entry.

Another way SMBs can benefit from LLM-powered tools is in their ability to plan. Every business collects data, whether on paper or on a computer.

For many small businesses, this data is little more than required record-keeping. However, with the assistance of LLMs, this data can be used to provide insights that can be used to guide future directions.

Imagine a business owner telling Microsoft Excel, "tell me how I should adjust my prices this year based on sales changes and inflation," rather than having to memorise a series of complex equations to get the same information.

With the power of large language models just becoming apparent, a whole new world is opening for SMBs. The Microsoft Africa Research Institute (MARI) is investigating how small businesses can best benefit from LLM-powered tools and systems.

We are designing and testing prototypes in real-world environments with the goal of creating tools and products that will significantly improve the quality of output for everyone.

The author is a Senior Applied Scientist at the Microsoft Africa Research Institute

PAYE Tax Calculator

Note: The results are not exact but very close to the actual.