Google DeepMind Blog ยท May 19, 2026

Introducing Gemini Omni

Why it matters

Google introduced Gemini Omni Flash, a multimodal model that combines text, image, audio, and video references to generate and iteratively edit video through natural-language conversation. Generated videos include a SynthID watermark.

My takeaway: Conversational video editing broadens impersonation, provenance, and policy risks. Deployments need identity protections, durable generation metadata, watermark verification, misuse testing, and review controls for externally distributed media.