NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild

TL;DR

The NTIRE 2026 Challenge focused on developing robust AI-generated image detection models capable of handling real-world transformations, using a large diverse dataset and evaluating performance with ROC AUC.

Contribution

This paper introduces a comprehensive challenge dataset and evaluation framework for detecting AI-generated images under various transformations, advancing robustness in practical scenarios.

Findings

Top methods achieved high ROC AUC scores on transformed images.

Diverse transformations significantly impact detection performance.

The challenge attracted over 500 participants, fostering progress in the field.

Abstract

This paper presents an overview of the NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild, held in conjunction with the NTIRE workshop at CVPR 2026. The goal of this challenge was to develop detection models capable of distinguishing real images from generated ones in realistic scenarios: the images are often transformed (cropped, resized, compressed, blurred) for practical usage, and therefore, the detection models should be robust to such transformations. The challenge is based on a novel dataset consisting of 108,750 real and 185,750 AI-generated images from 42 generators comprising a large variety of open-source and closed-source models of various architectures, augmented with 36 image transformations. Methods were evaluated using ROC AUC on the full test set, including both transformed and untransformed images. A total of 511 participants registered, with 20 teams submitting valid final solutions. This report provides a comprehensive overview of the challenge, describes the proposed solutions, and can be used as a valuable reference for researchers and practitioners in increasing the robustness of the detection models to real-world transformations.