DF26：我們再也無法分辨真假

摘要

我們提出 DF26，一個新穎的基準測試，用於偵測包含由近期文字轉影片與圖像轉影片模型所產生之全合成片段的 AI 生成影片。這些影片收錄單人公開演講情境，涵蓋直接對鏡頭錄製、官方聲明與攝影棚訪談——271 部真實影片與 2,420 部由七個現代影片模型生成的合成影片。針對 DF26 的研究顯示，人類在偵測 AI 生成影片的表現，以及最先進的深度偽造偵測器，皆接近隨機猜測。我們的結果凸顯當前評估協定的限制，並促使我們需要能明確衡量對現代生成模型分佈偏移之穩健性的基準測試。

English

We introduce DF26, a novel benchmark for detecting AI-generated videos containing fully synthetic clips produced by recent text-to-video and image-to-video models. The videos capture single-person public-speaking scenarios, spanning direct-to-camera recordings, official statements, and studio interviews - 271 real and 2,420 synthetic videos generated by seven modern video models. The study on DF26 shows that human performance in detecting AI-generated videos, as well as state-of-the-art deepfake detectors, is close to random chance. Our results highlight the limitations of current evaluation protocols and motivate the need for benchmarks that explicitly measure robustness to modern generative model distribution shifts.