Published
2025. 10. 10.
Author
ZAKE SHIM
Why don't my skills improve even though I'm using Midjourney?
Why don't my skills improve even though I'm using Midjourney?
1. Psychological barriers and the pursuit of efficiency.
When first encountering new technology, people tend to prefer "intuitive use." This means they want the process of using AI to be as simple and convenient as possible, leading to a desire to achieve results with minimal instructions.
This is a natural instinct for users to pursue efficiency. They often find complex inputs burdensome, so they tend to expect optimal results with minimal commands rather than complexly controlling AI.
2. Lack of language cognition and information organization skills.
In many cases, people lack sufficient literacy and information organization skills to effectively interact with AI, preventing them from writing detailed prompts.
Writing specific and clear prompts requires the ability to clearly define the desired outcome and express it in text.
However, these skills require continuous learning and training, and the public often tends to rely on short texts, vaguely expecting AI to do the work for them.
3. Cognitive burden and limitations of imagination. To utilize AI for creative work, users must visualize specific details and organize them into text.
Furthermore, the ability to imagine details and describe them in detail is a skill developed through personal training, so many people feel burdened and leave this task to AI.
To develop imagination, first cultivate observational skills.
To develop observational skills, first try to expose yourself to many images and videos.
By observing good images and videos, you will be able to distinguish and compare good from bad.
If you can distinguish and select between these, you are ready to use your imagination to control AI.
This is called "discernment."
*Discernment - The ability to distinguish and select between good and bad.
4. Expectation of immediate feedback and lack of experience.
There is often an underlying expectation that AI will automatically produce optimal results. This is the result of technological advancements, advertising, and public relations that have led to the perception of AI as a "completely autonomous tool."
New users often harbor vague expectations about AI and fall into the subjective interpretation that short texts can produce the desired results.
Furthermore, due to a lack of experience in writing AI prompts, many fail to perceive the "need for specific prompts." This is the result of a complex mix of psychological convenience, limited literacy, imaginative and cognitive burden, and excessive expectations of AI.
AI prompting is similar to inferring new images from a code composed of the user's intended text or images. Thus, iterative and multiple inferences can yield a large number of possible scenarios.
Of course, it is the user's responsibility to carefully select and refine the inferred images to suit their specific needs.
Gen-AI's most powerful feature, the freedom from "time and place" constraints, demands the user's imagination to select a creative theme, prepare preliminary materials for expression, compose a layout, select an art style, select an output medium, define aspect ratios, color themes, apply textures, and gather specialized terminology. This approach, based on the concept of synthesizing the desired visual, requires detailed text prompts to express the visuals in their minds.
For this reason, when users develop the discerning eye for selecting images and the ability to write prompts,
visual generation prompt engineering, which interacts and collaborates with AI, will become a powerful tool for establishing one's own competitiveness as a field that grows in direct proportion to the advancement of AI technology.
1. Psychological barriers and the pursuit of efficiency.
When first encountering new technology, people tend to prefer "intuitive use." This means they want the process of using AI to be as simple and convenient as possible, leading to a desire to achieve results with minimal instructions.
This is a natural instinct for users to pursue efficiency. They often find complex inputs burdensome, so they tend to expect optimal results with minimal commands rather than complexly controlling AI.
2. Lack of language cognition and information organization skills.
In many cases, people lack sufficient literacy and information organization skills to effectively interact with AI, preventing them from writing detailed prompts.
Writing specific and clear prompts requires the ability to clearly define the desired outcome and express it in text.
However, these skills require continuous learning and training, and the public often tends to rely on short texts, vaguely expecting AI to do the work for them.
3. Cognitive burden and limitations of imagination. To utilize AI for creative work, users must visualize specific details and organize them into text.
Furthermore, the ability to imagine details and describe them in detail is a skill developed through personal training, so many people feel burdened and leave this task to AI.
To develop imagination, first cultivate observational skills.
To develop observational skills, first try to expose yourself to many images and videos.
By observing good images and videos, you will be able to distinguish and compare good from bad.
If you can distinguish and select between these, you are ready to use your imagination to control AI.
This is called "discernment."
*Discernment - The ability to distinguish and select between good and bad.
4. Expectation of immediate feedback and lack of experience.
There is often an underlying expectation that AI will automatically produce optimal results. This is the result of technological advancements, advertising, and public relations that have led to the perception of AI as a "completely autonomous tool."
New users often harbor vague expectations about AI and fall into the subjective interpretation that short texts can produce the desired results.
Furthermore, due to a lack of experience in writing AI prompts, many fail to perceive the "need for specific prompts." This is the result of a complex mix of psychological convenience, limited literacy, imaginative and cognitive burden, and excessive expectations of AI.
AI prompting is similar to inferring new images from a code composed of the user's intended text or images. Thus, iterative and multiple inferences can yield a large number of possible scenarios.
Of course, it is the user's responsibility to carefully select and refine the inferred images to suit their specific needs.
Gen-AI's most powerful feature, the freedom from "time and place" constraints, demands the user's imagination to select a creative theme, prepare preliminary materials for expression, compose a layout, select an art style, select an output medium, define aspect ratios, color themes, apply textures, and gather specialized terminology. This approach, based on the concept of synthesizing the desired visual, requires detailed text prompts to express the visuals in their minds.
For this reason, when users develop the discerning eye for selecting images and the ability to write prompts,
visual generation prompt engineering, which interacts and collaborates with AI, will become a powerful tool for establishing one's own competitiveness as a field that grows in direct proportion to the advancement of AI technology.
Next Projects
