A prompt is a phrase formulated by humans when they communicate with the AI. In other words, when a person asks the AI to generate a piece of text or an image, what they do is enter the query, which essentially a prompt is. Prompts can be generalized, like asking for an image of a red cat, or extremely detailed, when every minor aspect of the output image is specified.
Although AI can understand words, sometimes — and you understand what we mean if you tried ChatGPT or Midjourney for yourself — it understands words differently than humans. This is the concern of prompt engineering or trying to find the optimal way of prompting in order to get the best results, i.e. those closest to the ones we, not the machines, have in our minds.
Generative AI refers to the subset of artificial intelligence focused on creating new content, ranging from images and music to text and more.
Text-to-Image Translation (T2I)
Text-to-image (T2I) translation is a type of artificial intelligence that generates an image based on a written description or a textual prompt.