Grok生成《奥德赛》短片遭多方批评，AI视频技术局限凸显

埃隆·马斯克旗下的Grok Imagine系统制作了一部时长七分钟的《奥德赛》介绍片，该片原计划作为完整长片的首集 1。然而，这部由人工智能生成的影片引发了媒体界的广泛批评，影评人指出其画面具有“恐怖谷”效应且表演缺乏灵魂 1。

在具体评价方面，《卫报》影评人彼得·布拉德肖（Peter Bradshaw）形容该片呈现出“恐怖谷般的库存素材质感”（uncanny-valley stock-footage quality） 1。《每日电讯报》则将其描述为“粗制滥造的血腥”（sloppy gruesomeness），而《滚石》杂志批评演员的表演显得“无灵魂”（soulless）和“僵硬”（wooden） 1。这些负面评价揭示了当前AI视频技术在叙事和情感表达上的显著短板。

尽管AI能够生成逼真的静态景观或带来意外的视觉美感，但在处理动态人物动作时难以保持镜头间的一致性，导致画面出现割裂感和塑料感 1。此外，AI无法像人类演员那样通过即兴发挥来创造真实的情感连接，文章以罗宾·威廉姆斯在《心灵捕手》中的表现为例说明了这一点 1。由于缺乏对人类经历和复杂情感的理解，AI目前仍难以创作出真正打动观众的叙事作品 1。值得注意的是，观众对已知由AI创作的 artworks 存在偏见，即使某些AI图像在未标明来源时可能在质量上优于人工作品 1。

Elon Musk’s Grok Imagine system generated a seven-minute introductory film for The Odyssey, planned as the first episode of a full-length feature 1. The project drew widespread media criticism, with reviewers arguing that despite its technical capabilities, it failed to evoke genuine emotion or human connection 1. Peter Bradshaw of The Guardian described the footage as having an “uncanny-valley stock-footage quality” 1. The Daily Telegraph labeled the visual content “sloppy gruesomeness,” while Rolling Stone magazine criticized the performances as “soulless” and “wooden” 1.

These critiques highlight specific technical limitations in current AI video generation tools. A major issue is the difficulty in maintaining consistency between shots when depicting human movement, often resulting in fragmented visuals and a plastic-like appearance 1. Furthermore, AI systems struggle to capture natural, complex emotions or assign logical motivations to characters, preventing them from creating narratives that truly resonate with audiences 1. Unlike human actors such as Robin Williams in Good Will Hunting, who use improvisation to forge authentic emotional connections, AI cannot replicate this spontaneity 1. Although Grok can produce realistic static landscapes or accidental aesthetic moments, the absence of an understanding of human experience limits its ability to craft moving stories 1. Additionally, viewer bias plays a role, as audiences tend to judge works known to be AI-generated more critically than similar human-made pieces, even if some AI images might technically be superior when their origin is not disclosed 1.