The Measurement of Artificial Intelligence (AI) Generative Accuracy via Text to Image: A Case Study of Thainess Perception of Digital Art

Artificial Intelligence (AI) is becoming popular and has had a huge impact in the digital art field. One problem is that the images generated by AI are sometimes not accurate in terms of reality. Therefore, the objective of this research is to 1) Study generative AI via text to image and Thainess. 2) Create an image from generative AI via text to image in terms of Thai perception 3) Measure generative AI via text to image to study Thai perception of digital art. The research methodology was divided into two phases. The first phase was to create 60 images using generative AI via text to image to determine Thai perception of digital art. The second phase was to measure the use of generative AI via text to image to study Thai perception of digital art. The sample consisted of five text to image AI generator platforms, all of which are well known in the digital art field, chosen by purposive sampling. The tools used in this research consisted of 60 images from Text-to-Image AI Generators, divided into 3 categories: landscape, portrait or object and abstract. The scoring rubric consisted of mean and standard deviation. The research results show that, Text-to-Image AI Generators can create an image which Thai individuals perceived as being at a good level for each of the 3 categories; landscape was perceived as being good, portrait or object as being satisfactory and abstract as being good.