Högskolan i Skövde

his.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Assessing human competence in distinguishing between human-created and AI-generated dragon artworks: A study on individuals’ ability to distinguish between AI-generated and human-created dragon artworks, focusing on art styles and prompt engineering techniques
University of Skövde, School of Informatics.
2024 (English)Independent thesis Basic level (degree of Bachelor), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

This study investigates the ability of individuals to distinguish between AI-generated and humancreated artworks, focusing on the influence of different art styles and the application of prompt engineering techniques. With advancements in AI enabling the creation of complex images, concerns about the authenticity and originality of art are increasing. The research addresses three main questions: the effectiveness of prompt engineering in mitigating inaccuracies in AI art generation, the public’s capacity to discern between AI-generated and human-created art, and the impact of various art styles on this discernment ability. Using DALL-E 3, a variety of dragon images were generated and evaluated for common inaccuracies. An online survey was conducted to collect data on participants’ perceptions and ability to identify AI-generated images across different art styles. The results indicate that prompt engineering can reduce inaccuracies. Respondents generally struggled to reliably distinguish between AI-generated and human-created art, with the art style influencing their discernment ability. The study underscores the complexities and potential biases in AI-generated content, highlighting the need for further research to enhance the reliability and authenticity of AI in the visual arts. 

The study was motivated by recent advancements in AI technologies that enable the creation of complex images and the controversies surrounding the use of AI in art, particularly in instances where AI-generated images are presented as human-made. To explore these issues, the research is structured around two main questions: whether people can accurately identify AI-generated images, and how different AI-influenced art styles affect their perceptions. 

The methodology includes prompt engineering techniques to generate images with specific characteristics, followed by a survey to collect data on how these images are perceived by a diverse audience. The survey assesses participants’ ability to distinguish between AI-generated and human-made images and records their opinions on the authenticity and appeal of different art styles influenced by AI. 

Place, publisher, year, edition, pages
2024. , p. 37
Keywords [en]
Generative AI, prompt engineering, art
National Category
Information Systems, Social aspects
Identifiers
URN: urn:nbn:se:his:diva-24280OAI: oai:DiVA.org:his-24280DiVA, id: diva2:1883112
Subject / course
Informationsteknologi
Educational program
Computer Science - Specialization in Systems Development
Supervisors
Examiners
Available from: 2024-07-09 Created: 2024-07-09 Last updated: 2024-07-09Bibliographically approved

Open Access in DiVA

fulltext(10753 kB)915 downloads
File information
File name FULLTEXT01.pdfFile size 10753 kBChecksum SHA-512
17d3fed9c37004ea159082bc7ba7ea158df27a3d74a355caa15aee2c1ee3bc7d5827582c1d89dcc468e3291147c1a228d1a7c7d2c8bdd3d9d903aa1d89aea7d1
Type fulltextMimetype application/pdf

By organisation
School of Informatics
Information Systems, Social aspects

Search outside of DiVA

GoogleGoogle Scholar
Total: 915 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 1410 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf