Concept

　I'm thinking, “Isn't this something interesting?” I am going to try to create various things based on the idea of Who is Generative AI?　Chat GPT has become famous, so is it something that chats? 　It seems to be able to create images and videos?　 I felt that it has the potential to do a lot of things, so I wondered if I could create some interesting content using generative AI.

　Generative AI has a base model, which can process text, images, or video. Among them, Chat GPT is one that processes text, Stable Diffusion is one that processes images, and Sora is one that processes video.

　Here, the content is created using the basic model of text and image generation AI. We intend to gradually add more and more content in the future. We hope you will enjoy experimenting with the AI and see how it responds or struggles to respond, depending on the words or images you input.

Contents

• How do you feel now? How do you feel?

　This content uses an underlying model of AI that generates “text” and “images”. The input is text, and the output is a text and an image. The input is “some words,” and the output is a text (text) and an image (a pictorial representation of the depth of your mood) based on the analysis of the depth of your mood. 　Enter “today,” “right now,” or words that describe your mood, and let the generative AI query and display your depth perception. For example, try entering a few words that describe how you are feeling right now, such as “I feel good,” “happy,” “happy,” “lonely,” or “sad. The generated AI will respond with an answer based on that information about how you are feeling today and right now. You can decide for yourself if you are right or not. It may express your state of mind that you are not aware of. Please enjoy what kind of answers you will get.

• What kind of landscape do you want to paint?

　　This content uses the base model of AI for generating “images”. The input is text and the output is an image. Simply put, this is a drawing content that “lets you draw a landscape” with the “words” you input. 　Try inputting words that point to the scenery you want the user to draw. It may be interesting to enter words that remind you of the scenery that comes to mind when you hear the word “scenery. Based on the inputted words, the AI will collect information and draw a landscape like a painting. Please enjoy what kind of scenery will be expressed and presented. Please feel free to give it a try.

• What kind of picture do you want to paint with words?

　　This content uses the base model of AI for generating “images”. The input is text and the output is an image. Simply put, this is a drawing content that “lets you draw a picture” with the input “words. This content is very similar to the second content above, but while the second content is limited to “scenery,” this content does not limit the subject of the picture. Therefore, it may be interesting to see how the picture-generating AI expresses a picture when the input is a reckless combination of words that cannot be fully drawn. 　The AI will draw a picture by making full use of information obtained from the input words. （Please enjoy the experience to see what kind of picture will be drawn (e.g., “dog sunglasses ball”). It may give you some hints or be useful in some way, so please feel free to give it a try.

• What kind of picture do you create from that picture?

　　This content uses the base model of AI to generate “text” and “images”. The input is an image (jpeg or png) and the output is also presented as an image. First, the “image” is presented as input, and the text is presented as output based on the analysis of the content of the input image. Next, the textual information resulting from the analysis of the image content is taken as input, and an image is presented as output based on the semantic content of this text. This is a somewhat unreasonable challenge for the AI to read the content from the picture. Try it out and see what kind of results you get. 　It may be interesting to compare the results of the AI with those of your own drawings, as well as with those of your favorites, pictures you are interested in, and so on. （It might be interesting to compare the results with those of the AI. It may give you some hints or be useful in some way, so please feel free to try it out.

• What psychological state does that picture indicate?

　　This content uses the basic model of AI for “text” generation. The input is an “image” (jpeg or png) and the output is presented as “text. The “image” is used as input, the image is psychologically analyzed, and the result is output as text. As with the previous contents, we are asking the AI to perform the difficult task of “reading the psychological state of the author from the picture”, which is somewhat impossible. Try it out and see what kind of results you get. 　It may be interesting to input (upload) not only your own pictures but also pictures (images) drawn by others and think about what kind of state of mind the creator was in when he/she created the picture. It may be interesting to compare the results of various drawings and see how they are expressed. Please feel free to give it a try.