Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Crafting Synthetic Realities: Examining Visual Realism and Misinformation Potential of Photorealistic AI-Generated Images

Qiyao Peng, Yingdan Lu, Yilang Peng, Sijia Qian, Xinyi Liu, Cuihua Shen

TL;DR

This study investigates photorealistic AI-generated images (AIGIs) and their misinformation potential by introducing a four-dimensional realism framework (content, human, aesthetic, AI production). It analyzes 4,335 photorealistic AIGIs drawn from a 30,824-image corpus collected from Instagram and Twitter using a mixed-methods approach (quantitative content analysis and qualitative review). Key findings show that AIGIs commonly feature humans, including politicians and celebrities, with high aesthetic professionalism and substantial surrealism, while overt AI-production signals are rare. The work highlights a high risk of misinformation due to realistic portrayals, low watermarking, and strong engagement drivers, proposing design interventions such as provenance labels (e.g., C2PA) and broader regulatory considerations to curb misuse and improve user discernment.

Abstract

Advances in generative models have created Artificial Intelligence-Generated Images (AIGIs) nearly indistinguishable from real photographs. Leveraging a large corpus of 30,824 AIGIs collected from Instagram and Twitter, and combining quantitative content analysis with qualitative analysis, this study unpacks AI photorealism of AIGIs from four key dimensions, content, human, aesthetic, and production features. We find that photorealistic AIGIs often depict human figures, especially celebrities and politicians, with a high degree of surrealism and aesthetic professionalism, alongside a low degree of overt signals of AI production. This study is the first to empirically investigate photorealistic AIGIs across multiple platforms using a mixed-methods approach. Our findings provide important implications and insights for understanding visual misinformation and mitigating potential risks associated with photorealistic AIGIs. We also propose design recommendations to enhance the responsible use of AIGIs.

Crafting Synthetic Realities: Examining Visual Realism and Misinformation Potential of Photorealistic AI-Generated Images

TL;DR

This study investigates photorealistic AI-generated images (AIGIs) and their misinformation potential by introducing a four-dimensional realism framework (content, human, aesthetic, AI production). It analyzes 4,335 photorealistic AIGIs drawn from a 30,824-image corpus collected from Instagram and Twitter using a mixed-methods approach (quantitative content analysis and qualitative review). Key findings show that AIGIs commonly feature humans, including politicians and celebrities, with high aesthetic professionalism and substantial surrealism, while overt AI-production signals are rare. The work highlights a high risk of misinformation due to realistic portrayals, low watermarking, and strong engagement drivers, proposing design interventions such as provenance labels (e.g., C2PA) and broader regulatory considerations to curb misuse and improve user discernment.

Abstract

Advances in generative models have created Artificial Intelligence-Generated Images (AIGIs) nearly indistinguishable from real photographs. Leveraging a large corpus of 30,824 AIGIs collected from Instagram and Twitter, and combining quantitative content analysis with qualitative analysis, this study unpacks AI photorealism of AIGIs from four key dimensions, content, human, aesthetic, and production features. We find that photorealistic AIGIs often depict human figures, especially celebrities and politicians, with a high degree of surrealism and aesthetic professionalism, alongside a low degree of overt signals of AI production. This study is the first to empirically investigate photorealistic AIGIs across multiple platforms using a mixed-methods approach. Our findings provide important implications and insights for understanding visual misinformation and mitigating potential risks associated with photorealistic AIGIs. We also propose design recommendations to enhance the responsible use of AIGIs.
Paper Structure (20 sections, 5 figures, 3 tables)

This paper contains 20 sections, 5 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Literature Review
  3. AIGI and AI Photorealism
  4. Understanding Photorealistic AIGIs through Perceived Realism
  5. Data and Methods
  6. Social Media Data
  7. Quantitative and qualitative analysis
  8. Results
  9. Quantitative Analysis of AIGI Features
  10. Human-related Surrealism
  11. Surreal Physical Characteristics
  12. Surreal Behaviors and Interactions
  13. Surreal Contexts
  14. Aesthetic Professionalism
  15. Discussion
  16. ...and 5 more sections

Figures (5)

  • Figure 1: Data Collection and Analytical Methods
  • Figure 2: AIGI examples of surrealism.
  • Figure 3: Staging and professionalism examples.
  • Figure 4: The Percentage of AIGI Visual Features
  • Figure 5: Examples of AI Production Flaw in AIGI