LAION-Mobile: Evaluating Deepfake Detectors On One Million Smartphone Photos

Authors: Achim von Stryk, Janis Keuper

Published: 2026-09-10 06:25:11+00:00

AI Summary

This paper introduces LAION-Mobile, a dataset of approximately one million smartphone images with EXIF metadata, to evaluate deepfake detectors on real-world computational photography. The study reveals that state-of-the-art deepfake detectors perform poorly on modern AI content (AUC no higher than 0.624) and exhibit high false-alarm rates (17-91%) on real smartphone photos when thresholds are calibrated on modern content. Consequently, no detector tested is both effective at identifying modern AI content and maintaining a deployable false-alarm rate on real images.

Abstract

Most Deepfake detectors report near-perfect AUC scores on their reference benchmarks. However, a recent ICML position paper argues that these evaluations collectively neglect the impact of modern smartphone photography: the widely used on-device neural image-signal processing pipelines (like multi-sensor fusion or noise and motion-blur suppression) increasingly shift the imaging paradigm from simple lens projections towards computational photography. Hence, devices actually generate, rather than record photos. This increases the risk that deepfake detectors may flag ordinary phone photos as fake. Due to the lack of large-scale datasets containing images from modern smartphones, this hypothesis has so far only been tested in small proof-of-concept studies. The aim of this paper is to close this gap. We introduce LAION-Mobile, an open dataset containing about 1 million smartphone images with EXIF metadata distilled from re-LAION-5B. Evaluating twelve state-of-the-art deepfake detectors with their original paper checkpoints on a 9,115-image evaluation sample of this pool (DIRE on 738), we report three key findings: (i) On modern AI content no detector exceeds AUC 0.624, and five of twelve fall below chance. (ii) Real-photo false-alarm rates are an artefact of threshold calibration: thresholds fitted on legacy GAN data make several detectors look deployable (less than 11 percent FPR), yet the same detectors flag 17-91 percent of real photos once the identical criterion is refit on modern content. (iii) Consequently, no detector both beats chance on modern AI content and keeps a deployable real-photo false-alarm rate. Mirroring the device mix of web collections, the corpus probes the first neural-ISP generation (2018-2020); current flagships are essentially absent, leaving the modern-ISP regime as the open gap.