YouTube: 2 new announcements from Google I/O

Source: Mashable Tech

Word of the day: Tokenmaxxing – the term Google CEO Sundar Pichai used to introduce this year’s Google I/O conference, referencing the massive AI processing (quadrillions of operations) happening worldwide.

Google has made a full pivot to AI, and across its many products the impact varies. Among the dozens of AI model updates and new multimodal video‑generation tools announced, Google highlighted a few upgrades for YouTube, the world’s most‑watched video platform.

For a complete rundown of Google I/O announcements, follow the Mashable Google I/O live blog for updates on Gemini, Chrome, Android, and more.

YouTube Shorts get Gemini Omni

Google introduced its new AI world model, Gemini Omni, early in the event, noting its two‑directional multimodal capabilities that can “create anything from any output.”

Omni will be integrated into YouTube Shorts Remix, a tool that generates videos from existing online content. With Omni, creators can remix Shorts using advanced AI prompts. Videos produced with Omni will automatically display an AI‑generated content label and associated metadata, linking back to the original source.

YouTube is also expanding its likeness detection tool, which helps creators identify content where their face has been altered or generated by AI, to all creators aged 18 +.

Ask YouTube: New way to search

The second YouTube‑related update is a conversational search feature that surfaces relevant YouTube videos directly within Google Search results. When users ask specific or complex questions—e.g., “how to teach my child to ride a bike”—they can watch tutorials or related videos right from the search page and interact with a tailored response generated by Google’s AI Mode.

Google describes the feature as a complete reimagining of the search experience. It is currently in testing and is slated for a broad rollout across the U.S. this summer.