Toward Accountable AI-Generated Content on Social Platforms: Steganographic Attribution and Multimodal Harm Detection
Authors: Xinlei Guan, David Arosemena, Tejaswi Dhandu, Kuan Huang, Meng Xu, Miles Q. Li, Bingyu Shen, Ruiyang Qin, Umamaheswara Rao Tida, Boyang Li
Organizations: Kean University, Union, NJ 07083 USA · North Dakota State University, Fargo, ND 58102 USA · McGill University, Montréal, QC H3A 0G4, Canada · the University of Notre Dame, Notre Dame, IN 46556 USA · Villanova University, Villanova, PA 19085 USA
The rapid growth of generative AI has introduced new challenges in content moderation and digital forensics. In particular, benign AI-generated images can be paired with harmful or misleading text, creating difficult-to-detect misuse. This contextual misuse undermines the traditional moderation framework and complicates attribution, as synthetic images typically lack persistent metadata or device signatures. We introduce a steganography enabled attribution framework that embeds cryptographically signed identifiers into images at creation time and uses multimodal harmful content detection as a trigger for attribution verification. Our system evaluates five watermarking methods across spatial, frequency, and wavelet domains. It also integrates a CLIP-based fusion model for multimodal harmful-content detection. Experiments demonstrate that spread-spectrum watermarking, especially in the wavelet domain, provides strong robustness under blur distortions, and our multimodal fusion detector achieves an AUC-ROC of 0.99, enabling reliable cross-modal attribution verification. These components form an end-to-end forensic pipeline that enables reliable tracing of harmful deployments of AI-generated imagery, supporting accountability in modern synthetic media environments. Our code is available at GitHub: https://github.com/bli1/steganography
Figures & tables
Fig. 1 : Overview of the Steganographic Security Gateway.
Fig. 2 : Flowchart of the Steganographic Enabled Image Tracing Pipeline.
Method
Total
Gaussian Blur
JPEG Compression (Q=50)
Resize
Success
Failure
Success
Failure
Success
Failure
Avg
Std
Rate(%)
Avg
Std
Avg
Std
Rate(%)
Avg
Std
Avg
Std
Rate(%)
Avg
Std
LSB
1100
1100.0
0.0
100.0
0.0
0.0
0.0
0.0
0.0
1100.0
0.0
1100.0
0.0
100.0
0.0
0.0
DCT
1100
779.7
2.5
70.9
320.3
2.5
0.0
0.0
0.0
1100.0
0.0
784.1
5.3
71.3
315.9
5.3
DWT
1100
795.0
3.2
72.3
305.0
3.2
0.0
0.0
0.0
1100.0
0.0
793.8
3.6
72.2
306.2
3.6
Spatial SS
1100
1086.8
1.7
98.8
13.2
1.7
1065.0
1.8
96.8
35.0
1.8
1085.8
3.5
98.7
14.2
3.5
TABLE I: Steganography Robustness Performance Across Different Attack Methods.
Fig. 3: Visual comparison of watermarking methods on a natural color image under different attack conditions.
Fig. 4 : Training Accuracy and Validation Performance.
With the rapid advancement of AIGC technologies, image forensics will encounter unprecedented challenges. Traditional methods are incapable of dealing with increasingly realistic images generated by rapidly evolving image generation techniques. To facilitate the identification of AI-generated images and the attribution of their source models, generative image watermarking and AI-generated image attribution have emerged as key research focuses in recent years. However, existing methods are model-dependent, requiring access to the generative models and lacking generality and scalability to new and unseen generators. To address these limitations, this work presents a new paradigm for AI-generated image attribution by formulating it as an instance retrieval problem instead of a conventional image classification problem. We propose an efficient model-agnostic framework, called Low-bIt-plane-based Deepfake Attribution (LIDA). The input to LIDA is produced by Low-Bit Fingerprint Generation module, while the training involves Unsupervised Pre-Training followed by subsequent Few-Shot Attribution Adaptation. Comprehensive experiments demonstrate that LIDA achieves state-of-the-art performance for both Deepfake detection and image attribution under zero- and few-shot settings. The code is at https://github.com/hongsong-wang/LIDA
Hongsong Wang, Renxi Cheng, Chaolei Han +1
School of Computer Science and Engineering, Southeast University, Nanjing 210096, China · Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications (Southeast University), Ministry of Education, China · School of Cyber Science and Engineering, Southeast University, Nanjing 210096, China +2
AI-generated image (AIGI) attribution presents a pressing challenge that goes beyond mere AIGI detection, aiming to identify the source model or technique responsible for a synthetic image. However, most previous source attribution methods operate in a closed-set manner, which necessitates retraining to recognize any novel category, preventing adaptation to the rapid evolution of image generation. In this work, we propose a new paradigm for synthetic image attribution, termed few-shot attribution. This paradigm targets the reliable identification of unseen generators using only limited samples, making it highly suitable for real-world applications. To facilitate this work, we construct OmniFake, a large-scale, well-categorized synthetic image dataset that contains 1.17 million images from 45 distinct generators. We further introduce OmniDFA (Omni Detector and Few-shot Attributor), a few-shot attribution baseline that not only assesses the authenticity of images but also determines their synthesis origins. Experiments demonstrate that OmniDFA exhibits excellent capability in few-shot attribution and achieves state-of-the-art generalization performance in AIGI detection. Our dataset and code are available at https://github.com/teheperinko541/OmniDFA.
Shiyu Wu, Shuyan Li, Jing Li +2
Institute of Automation, Chinese Academy of Sciences, Beijing, China · Beijing Academy of Artificial Intelligence, Beijing, China · University of Chinese Academy of Sciences, Beijing, China +3
The rapid advancement of generative models has blurred the boundary between synthetic and real imagery, creating an urgent need for reliable deepfake detection. Yet most existing approaches rely on massive real--fake datasets, which are increasingly difficult to maintain as new generators continue to emerge. In this work, we investigate how much information about image authenticity is already encoded in modern multimodal vision representations. We find that frozen multimodal encoders naturally separate real and synthetic images in their embedding space, enabling a simple linear classifier to achieve strong performance without task-specific fine-tuning. Motivated by this observation, we develop a representation-aware data curation strategy that selects a compact set of representative generators for training. The resulting training set contains only 10K images, compared to 288K in AIGIBench and 4M in OpenFake, while improving robustness to unseen generators and distribution shifts. We additionally introduce RealWorldBench, a benchmark consisting of modern camera photographs, contemporary stock images, and outputs from recent commercial generators. Experiments across multiple benchmarks show that combining frozen multimodal representations with carefully curated training data provides a simple and effective approach to AI-generated image detection.