Why prompting (Prompting engineering) in AI

In 1997, the transformer was introduced by Vaswani(“Attention Is All You Need”), which made a new paradigm and revolutionized various language processing tasks.

So, instead of conventional AI with a dedicated model for the specific task, the foundation model was considered for answering all types of questions. But as the biggest disadvantage of the foundation model, it really can’t answer up-to-date, contextual things based on your own company’s business.

Therefore, there are commonly two technologies to be utilized to unblock those constraints. One is ‘fine-tuning’ and the other one is ‘prompting’. The ways of doing ‘fine-tuning’ are various, however, training with the relevant datasets is one of the best solutions to make your model specialized for your business and tasks.

What is prompt engineering?

Prompt engineering refers to the process of designing and refining prompts or instructions given to a language model like ChatGPT to elicit a desired response.

For example, let’s say that I want to have the next word as ‘a book’ after the sentence ‘the student opens…’. (Tested with…