AI-generated images (AIGIs) have reached a level of quality at which humans are unable to reliably distinguish them from real images. To counter the inherent risks of fraud and disinformation, their detection is a pressing challenge and an active research topic. While many of the presented methods claim high detection accuracy, they are often evaluated without considering their adversarial robustness. In real-world scenarios, malicious actors can modify AIGIs, e.g., through imperceptible perturbations, to evade detection systems. In this work, we provide a dataset to assess the robustness of AI-generated image detectors. We present RAID (Robust evaluation of AI-generated image Detectors), a benchmark dataset of 216k diverse and transferable adversarial images. The proposed dataset is created starting from images generated by four different text-to-image models (to promote diversity) and optimizing attacks against an ensemble of seven state-of-the-art detectors (to promote transferability). Extensive experiments show that our methodology generates adversarial images that transfer to unseen detectors, degrading the performance of every detector outside the attack ensemble that we evaluate, and provides a practical, worst-case robustness screen: failure on RAID is a strong indicator that a detector is not adversarially robust.
RAID: A benchmark dataset for testing the adversarial robustness of AI-generated image detectors
Eddoubi, Hicham
Primo
;Sotgiu, Angelo;Pintor, Maura;Biggio, BattistaUltimo
2027-01-01
Abstract
AI-generated images (AIGIs) have reached a level of quality at which humans are unable to reliably distinguish them from real images. To counter the inherent risks of fraud and disinformation, their detection is a pressing challenge and an active research topic. While many of the presented methods claim high detection accuracy, they are often evaluated without considering their adversarial robustness. In real-world scenarios, malicious actors can modify AIGIs, e.g., through imperceptible perturbations, to evade detection systems. In this work, we provide a dataset to assess the robustness of AI-generated image detectors. We present RAID (Robust evaluation of AI-generated image Detectors), a benchmark dataset of 216k diverse and transferable adversarial images. The proposed dataset is created starting from images generated by four different text-to-image models (to promote diversity) and optimizing attacks against an ensemble of seven state-of-the-art detectors (to promote transferability). Extensive experiments show that our methodology generates adversarial images that transfer to unseen detectors, degrading the performance of every detector outside the attack ensemble that we evaluate, and provides a practical, worst-case robustness screen: failure on RAID is a strong indicator that a detector is not adversarially robust.| File | Dimensione | Formato | |
|---|---|---|---|
|
1-s2.0-S1077314226003693-main.pdf
accesso aperto
Descrizione: Pubblicazione in formato PDF rilasciata in open access dalla rivista.
Tipologia:
versione editoriale (VoR)
Dimensione
2.99 MB
Formato
Adobe PDF
|
2.99 MB | Adobe PDF | Visualizza/Apri |
I metadati presenti in IRIS UNICA sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono protetti da diritto d'autore, salvo diversa indicazione.



