Findings of the Counter Turing Test: AI-Generated Image Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Roy, Rajarshi, Imanpour, Nasrin, Aziz, Ashhar, Bajpai, Shashwat, Singh, Gurpreet, Biswas, Shwetangshu, Wanaskar, Kapil, Patwa, Parth, Ghosh, Subhankar, Dixit, Shreyas, Pal, Nilesh Ranjan, Rawte, Vipula, Garimella, Ritvik, Das, Amitava, Sheth, Amit, Sharma, Vasu, Reganti, Aishwarya Naresh, Jain, Vinija, Chadha, Aman
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911715180937216
author Roy, Rajarshi
Imanpour, Nasrin
Aziz, Ashhar
Bajpai, Shashwat
Singh, Gurpreet
Biswas, Shwetangshu
Wanaskar, Kapil
Patwa, Parth
Ghosh, Subhankar
Dixit, Shreyas
Pal, Nilesh Ranjan
Rawte, Vipula
Garimella, Ritvik
Das, Amitava
Sheth, Amit
Sharma, Vasu
Reganti, Aishwarya Naresh
Jain, Vinija
Chadha, Aman
author_facet Roy, Rajarshi
Imanpour, Nasrin
Aziz, Ashhar
Bajpai, Shashwat
Singh, Gurpreet
Biswas, Shwetangshu
Wanaskar, Kapil
Patwa, Parth
Ghosh, Subhankar
Dixit, Shreyas
Pal, Nilesh Ranjan
Rawte, Vipula
Garimella, Ritvik
Das, Amitava
Sheth, Amit
Sharma, Vasu
Reganti, Aishwarya Naresh
Jain, Vinija
Chadha, Aman
contents The rapid advancements in generative AI technologies, such as Stable Diffusion, DALL-E, and Midjourney, have significantly transformed the creation of synthetic visual content. While these models enable innovation across industries, they also pose serious challenges, including misinformation, disinformation, and biased content generation. The increasing realism of AI-generated images makes their detection a pressing concern for researchers, policymakers, and industry stakeholders. In this paper, we present the findings of the Defactify 4.0 workshop, which introduced the Counter Turing Test (CT2) for AI-Generated Image Detection. The competition consisted of two key tasks: (1) binary classification of images as either AI-generated or real and (2) identification of the specific generative model responsible for an AI-generated image. To support both tasks, we employed the MS COCOAI dataset, a benchmark of 96000 real and synthetic images generated by five state-of-the-art models alongside real images from MS COCO. Participants employed diverse detection strategies, including convolutional neural networks (CNNs), Vision Transformers (ViTs), frequency-based analysis, contrastive learning, and multimodal techniques. The results demonstrated that while AI-generated images can be detected with high accuracy (F1-score > 0.83), identifying the exact model used remains significantly more challenging (highest F1-score: 0.4986). These findings highlight the need for improved model fingerprinting, adversarial robustness, and real-time detection mechanisms.
format Preprint
id arxiv_https___arxiv_org_abs_2605_20787
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Findings of the Counter Turing Test: AI-Generated Image Detection
Roy, Rajarshi
Imanpour, Nasrin
Aziz, Ashhar
Bajpai, Shashwat
Singh, Gurpreet
Biswas, Shwetangshu
Wanaskar, Kapil
Patwa, Parth
Ghosh, Subhankar
Dixit, Shreyas
Pal, Nilesh Ranjan
Rawte, Vipula
Garimella, Ritvik
Das, Amitava
Sheth, Amit
Sharma, Vasu
Reganti, Aishwarya Naresh
Jain, Vinija
Chadha, Aman
Computer Vision and Pattern Recognition
The rapid advancements in generative AI technologies, such as Stable Diffusion, DALL-E, and Midjourney, have significantly transformed the creation of synthetic visual content. While these models enable innovation across industries, they also pose serious challenges, including misinformation, disinformation, and biased content generation. The increasing realism of AI-generated images makes their detection a pressing concern for researchers, policymakers, and industry stakeholders. In this paper, we present the findings of the Defactify 4.0 workshop, which introduced the Counter Turing Test (CT2) for AI-Generated Image Detection. The competition consisted of two key tasks: (1) binary classification of images as either AI-generated or real and (2) identification of the specific generative model responsible for an AI-generated image. To support both tasks, we employed the MS COCOAI dataset, a benchmark of 96000 real and synthetic images generated by five state-of-the-art models alongside real images from MS COCO. Participants employed diverse detection strategies, including convolutional neural networks (CNNs), Vision Transformers (ViTs), frequency-based analysis, contrastive learning, and multimodal techniques. The results demonstrated that while AI-generated images can be detected with high accuracy (F1-score > 0.83), identifying the exact model used remains significantly more challenging (highest F1-score: 0.4986). These findings highlight the need for improved model fingerprinting, adversarial robustness, and real-time detection mechanisms.
title Findings of the Counter Turing Test: AI-Generated Image Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.20787