How Useful Are LLMs In Business?

Continuing LLMs: How Useful Are LLMs In Business?
Large language models have become very useful. They allow you to type or use voice control to ask a question. It hardly matters how you phrase the question, or if your grammar or spelling is suspect, the LLM usually is able to understand. This is because it has so much knowledge, so much “experience” that it invariably gets it right. Similarly its answers are mostly spot-on.
In a way it is seen to be able to create new knowledge, but only because it has so much data at hand. That’s why it can “write a best man’s speech in the style of Shakespeare” and the speech is pretty good. All it is doing is joining up various words and phrases in a way that makes sense, but it is very impressive.
In business, it can help with any type of writing or response. “Write an email thanking President Trump for inviting me to his inauguration, but decline the invitation in a very respectful way”
Try it. It’s very impressive. If you did receive an invitation, ChatGPT would have saved you hours. It can do the same for almost any task.

The LLM knowledgebase is huge and getting bigger.
There’s another better use of the LLM.
This involves taking away its huge set of answers and substituting it with your own.
What???? Why would I want to do that???

We want to retain its ability to understand and reply, but modify its reply only for your purposes. In this scenario the set of answers consists of your company knowledge. Obviously this dataset is confidential and is only available to people you authorise. It can be staff that have access to HR issues, it can be customer support staff that have access to customer data, it can be customers that have access to product information.

In this case the AI LLM is able to extract relevant data from your company, keep it confidential, and answer the person with the right type of information. Imagine an employee wanting to know about a grievance procedure, or a customer choosing to speak Afrikaans, or a director wanting a summary of the latest annual financial statements and then to query specific items. The AI LLM is better able to interrogate the data and turn it into useful knowledge completely automatically.