top of page

BusinessDay: Large language models — the key to unlocking corporate data

The latest artificial intelligence tools can transform how internal data is accessed and presented.

By Johan Steyn, 6 December 2023

In the corporate world, it is often the case that vast repositories of information are spread across numerous systems. The challenge of finding accurate and relevant information from various systems in corporations has become a significant hurdle, affecting efficiency, decision-making, and ultimately, the performance of internal staff.

One of the primary issues facing large corporations is the existence of multiple, often siloed, information systems. Ranging from internal databases and document management systems to customer relationship management platforms and enterprise resource planning software, these systems are typically designed to serve specific functions. As a result, crucial information is scattered, making it difficult for employees to locate and access the data they need efficiently.

The sheer volume of data housed in these systems adds another layer of complexity. Large corporations generate and store an enormous amount daily, encompassing everything from internal communications and project reports to market research and financial records. Sifting through this avalanche to find specific, actionable information is often akin to finding a needle in a haystack.

I have recently worked on a case study with a client’s human resources (HR) department. If I wanted to take paternity leave in January, as an example, I would typically have to access a number of information portals. First, I would have to try to find the company’s leave policy document. Then I need to see what the labour law says about such leave. I then need to determine how many annual leave days I could use at that time to extend the days away from work. Lastly I would need to confirm if I have any work commitments booked for that time, such as project work.

It would take me some time to ultimately find an answer to a simple question. What if I could just ask the question in the HR portal without having to navigate various systems? Just imagine that the portal could quickly access all the relevant sources and provide me with the answer I was looking for.

The good news for business leaders is that the latest arsenal of artificial intelligence (AI) tools can transform how internal data is accessed across multiple systems and presented in a meaningful way. Large language models (LLMs), central to platforms such as OpenAI’s ChatGPT and Google’s Bard, are sophisticated AI models developed to comprehend, process, and produce natural language akin to human speech. Capable of executing a diverse array of linguistic tasks, these models excel in activities such as creating content, summarising text, translating languages and answering questions.

We are now witnessing a significant shift in the way we access information. In the era we are entering traditional methods of search are being transformed into more dynamic and intuitive processes of information discovery. This evolution represents a fundamental change in how we interact with data, mirroring the way human curiosity and understanding naturally unfold.

Business leaders should consider the integration of LLMs into their corporate systems. The key is to start by identifying specific areas in their organisations where information retrieval challenges are most pronounced. By staying informed about advancements in AI and adapting their approaches accordingly, business leaders can ensure that their teams remain at the forefront of efficient data management and decision-making processes.


bottom of page