The video generation and editing paradigm is shifting from tedious timeline-based workflows to a few casual sentences with an AI. At Google I/O 2026 (May 19), the company unveiled Gemini Omni, its first "world model" that natively blends Gemini's reasoning intelligence with media generation. The first version, Omni Flash, takes multimodal inputs — text, images, video, audio — and produces high-fidelity video or allows conversational editing. Unlike earlier models like Sora, which only...
Tags: #Gemini