A leak of Google's upcoming AI video tool, potentially named Veo 4 or Gemini Omni, suggests a significant advancement in AI video generation. The tool reportedly enables the creation of multi-angle scenes with smooth perspective transitions and synchronized audio, including dialogue and ambient sounds. This development marks a shift from AI as a tool for short clips to a comprehensive narrative production tool, capable of generating up to 9-second videos at 720p resolution. The leak, revealed by Pankaj Kumar, indicates that Veo 4 could overcome previous limitations in AI video production, such as the inability to cut scenes while maintaining continuity. This advancement positions AI as a "director" rather than just a "painter," capable of handling complex narrative tasks. The leak also highlights Google's strategic timing, following the shutdown of OpenAI's Sora app, which struggled with high costs and low user retention. Google's I/O event is expected to reveal more about Veo 4 and other Gemini models, potentially reshaping the AI video landscape.