The paper evaluates whether multimodal large language models (MLLMs) can serve as forensic detectors of deepfake facial images. Six MLLMs—including GPT-4o, GPT-5, GPT-5.2, Gemini 2.5, Gemini 3, and Seed 1.6—were tested on a dataset of 400 images (50 authentic and 350 synthetic) generated by seven leading image models. Results show that several MLLMs achieve high detection accuracy, with GPT-5.2 reaching 97.9% in pairwise comparisons, but the presence of invisible SynthID watermarks can reduce detection performance by about 10% for Gemini.
arXiv:2608. 06865v1 Announce Type: cross Abstract: The malicious use of generative artificial intelligence to create highly realistic deepfake videos raises serious ethical concerns and poses substantial challenges to AI safety.
arXiv:2609.01511v1 Announce Type: new Abstract: Face forgery detectors often achieve strong results on controlled benchmarks, but their reliability under realistic image degradations remains limited....
arXiv:2609.13617v1 Announce Type: new Abstract: We evaluate whether the improvements advertised for ChatGPT Images 2.5 translate into better performance on forgery tasks with predetermined answers. W...
The paper proposes Artifact-Complementary Expert Fusion (ACEF), a two‑stage framework that enhances AI‑generated image detection by combining two types of reconstruction artifacts—VAE/DDIM and SRGAN—into aligned synthetic negatives. ACEF first builds artifact‑specific experts using LoRA adaptation on a frozen backbone, then fuses their multi‑layer evidence with Layer‑wise Artifact‑Complementary Fusion (LACF) to mitigate conflicts between artifact manifolds. Experiments on 13 benchmarks show that this approach improves generalizability over existing state‑of‑the‑art methods.
arXiv:2607. 06615v1 Announce Type: cross Abstract: Image forgery detection is a critical task in digital forensics, yet many deep-learning localization approaches are typically GPU-accelerated and computationally heavier than handcrafted screening methods.
arXiv:2607. 13234v1 Announce Type: cross Abstract: Deepfake detectors that achieve near-perfect scores on academic benchmarks collapse on real-world content: recent in-the-wild evaluations report AUC drops of 45-50% for state-of-the-art open-source models.
FARE (Forensic Acceptance Region Estimation) is a method designed to audit the integrity of AI image generators at deployment time. By training on images from a certified generator, FARE learns generator‑specific forensic features and tightens its acceptance region using hard samples, enabling it to determine whether a new image originates from the enrolled generator. Experiments show that FARE consistently outperforms existing baselines in detecting generator swaps, even when the swapped model is similar or under exact‑model and decision‑only attacks.
arXiv:2609.37783v1 Announce Type: cross Abstract: Photographic evidence is becoming increasingly vulnerable to forms of alteration and fabrication that existing legal and technical workflows are not...
ManiVid introduces a unified forensic analysis framework for manipulated videos, combining forgery detection, artifact grounding, and anomaly explanation. The authors release ManiVid-38K, a large dataset of 19K real‑fake video pairs with authenticity labels, forgery masks, and explanations, and a benchmark ManiVidBench with 1K balanced pairs. ManiVidLens, the proposed model, outperforms existing methods in artifact grounding and anomaly explanation while matching state‑of‑the‑art detection accuracy.
arXiv:2608. 04840v1 Announce Type: cross Abstract: Verifying the authenticity of satellite imagery has become increasingly critical given advances in generative artificial intelligence.
The paper introduces TaintedPixels, a proactive video‑protection technique that embeds subtle, structured perturbations into the blue channel of facial regions. These perturbations remain invisible in the original video but become noticeable after black‑box manipulation, thereby flagging deepfakes. Experiments on three off‑the‑shelf manipulation tools and two detectors show that TaintedPixels achieves the highest forgery detection rate while keeping visual distortion minimal (LPIPS = 0.0042), and a human study confirms that protected videos are rarely suspected, whereas forgeries from protected sources are identified as fake far more often than those from unprotected sources.
arXiv:2607. 22745v1 Announce Type: cross Abstract: Rapid advances in image generation are eroding the evidentiary value of visual content in settings where authenticity can affect public safety and personal reputation.