Photo credit: www.gadgets360.com
YouTube has unveiled a new artificial intelligence (AI) functionality for its creators, enhancing the capabilities of its platform. Announced on Thursday, this update sees the integration of Google’s Veo 2 AI model with the existing Dream Screen feature. This advancement allows users not only to enhance their videos with AI-generated backgrounds but also to create entirely new standalone videos from scratch. The platform aims to provide creators with tools to visualize concepts that may be challenging to capture through traditional means.
YouTube Shorts Now Allows Users to Generate AI Videos
In a recent blog post, YouTube emphasized the collaboration with Google DeepMind to incorporate the Veo 2 video generation model into its Dream Screen feature. This addition facilitates the creation of standalone videos specifically tailored for YouTube Shorts, the platform’s short-form vertical video format introduced in 2020.
The newly launched feature is designed to assist content creators in sourcing footage that is otherwise difficult to obtain. Acting as a text-to-video generator, users need only to describe their desired content, and the Veo 2 model will produce a matching video output.
The Veo 2 model, which debuted in December 2024 as an upgrade to its predecessor, has been touted for its significant enhancements in video quality, detail, and realism. According to DeepMind, this AI model demonstrates a superior understanding of cinematography, encompassing various genres, lens types, cinematic effects, and camera motions.
To utilize the Veo 2 for video creation, users must access the Shorts camera, select Add to initiate the media picker, and then tap Create at the top of the interface. This process leads to a text input field where users can input their prompts. Following this, they can choose a stylistic option, click on Create video, and determine the desired length of their video.
To address concerns regarding potential misuse of AI in creating misleading content, YouTube plans to implement SynthID watermarks on all AI-generated videos. This feature is currently limited to specific regions, including Australia, Canada, New Zealand, and the United States, with further expansion anticipated in the future.
Source
www.gadgets360.com