gen‑ai.news
← Back
Video

Google announces new 'Video Remix' feature its for AI subscribers

Google announces new 'Video Remix' feature its for AI subscribers

Google has announced Video Remix, a new feature available to its AI subscription subscribers that brings generative video editing to Google Photos. Powered by Gemini Omni, the tool allows users to take videos already stored in their Photos library and reimagine them in new ways - applying stylistic transformations or other AI-driven modifications to existing footage.

The feature is notable because it targets personal, real-world video rather than purely synthetic generation. Instead of prompting an AI to create something from nothing, users are working with their own memories and recorded moments, using Gemini Omni as a kind of creative filter or reinterpretation engine. This positions Video Remix closer to an editing assistant than a standalone generation tool.

Gemini Omni, Google's most capable multimodal model family, underpins the feature. Its ability to understand and process video content at a detailed level is what makes context-aware remixing possible, rather than simple surface-level effects. Google has been steadily integrating Gemini across its suite of apps and services, and Video Remix continues that pattern by embedding the model into one of its most widely used consumer platforms.

The feature is tied to Google's AI subscription, meaning it sits behind a paywall rather than being available to all Photos users. This approach mirrors how competitors like Adobe and Apple have structured access to their more advanced AI editing tools. For subscribers, Video Remix adds a practical creative outlet that works directly with content they already own and have stored in Google's ecosystem.

Read at Engadget →
Share:X

Enjoy this story? Get the next one in your inbox.

Twice a week: the most important stories in generative image and video AI, distilled into a 2-minute read.

Free. Unsubscribe any time. No spam, ever.

Your next read

World models that ignore human beliefs predict the wrong actions, new research shows
Video

World models that ignore human beliefs predict the wrong actions, new research shows

Current AI world models simulate physical environments but leave out a critical layer: what people believe, want, and intend. New research introduces a "Mental World Modeling" framework that adds these mental variables, and finds that even smaller models using it can outperform larger ones that ignore human psychology. The key bottleneck turns out to be modeling how physical and mental states evolve together over time.

Runway News | The Next Phase of Enterprise Video Generation
Video

Runway News | The Next Phase of Enterprise Video Generation

Runway's Chief Revenue Officer has distilled hundreds of enterprise conversations into five themes shaping how large organizations are approaching AI video generation. The piece covers everything from model consolidation and data sovereignty to shifting cost structures and the move toward autonomous execution.

Major YouTube creators are facing backlash for accepting AI money
Video

Major YouTube creators are facing backlash for accepting AI money

Several prominent filmmaking YouTubers, including Matti Haapoja and Sam "Kold" Kolder, have drawn criticism after posting sponsored content promoting Higgsfield's AI video platform without clearly disclosing the paid nature of those partnerships. The backlash intensified when other creators began sharing apparent screenshots of outreach from PR firms working on Higgsfield's behalf. The episode has sparked a broader conversation about transparency and trust in the creator community around AI tool