YouTube Shorts Integrates Google's Veo 2 AI for Advanced Video Generation

YouTube Rolls Out AI-Generated Green Screens for Shorts

YouTube Shorts has rolled out more generative AI features for its Dream Screen program which is based on Google's Veo 2 video model. Dream Screen was introduced in September 2023 and the app already supports generating AI images and videos for Shorts based on written prompts. But the latest updates allow creators to AI-generate virtual green screen backgrounds. It is also now powered by Google DeepMind's newest video generation model, Veo 2. "Need a specific scene but don't have the right footage? Want to turn your imagination into reality and tell a unique story? Simply use a text prompt to generate a video clip that fits perfectly into your narrative, or create a whole new world of content. It's that easy!," YouTube writes in a blog update. YouTube says that Veo 2 creates "state-of-the-art, high-quality videos in a wide range of subjects and styles that better match your vision." It also can create AI content faster than before. "Veo 2 understands real-world physics and human movement better, making its output more detailed and realistic. You can even specify a style, lens, or cinematic effect, making Dream Screen an easy and fun way to express yourself," it adds. The creative example above shows one YouTuber enlarging his pomeranian for a comedic skit. YouTube says that to create video backgrounds for your Shorts, open the Shorts camera, select Green Screen, then Dream Screen. From there, type in a prompt, select your image, and then create your video background. To create a clip to add to any Short, open the Shorts camera, tap Add to open the media picker, then tap Create at the top. After inputting your prompt, select your image, then tap Create video, and choose your desired length. "We use SynthID watermarks and clear labels to indicate that these creations were generated with AI," adds YouTube. "These features are launching now for anyone in the US, Canada, Australia, and New Zealand, with plans to expand to more of you in the future."

YouTube Shorts is getting a tool for adding AI-generated video to your posts

It's about to get easier to add AI-generated video into your short-form posts. YouTube announced that Shorts will integrate Veo 2 -- Google DeepMind's newest video mode -- which will allow users to generate and add AI clips into their videos. The new feature will be a part of Shorts' Dream Screen, which already lets users add AI-generated backgrounds to their posts. Wrote Youtube in a blog post announcing the news: "Need a specific scene but don't have the right footage? Want to turn your imagination into reality and tell a unique story? Simply use a text prompt to generate a video clip that fits perfectly into your narrative, or create a whole new world of content. It's that easy!" YouTube claimed in the blog post that Veo 2 better "understands real-world physics and human movement," which should result in more detailed and accurate results. Users should also be able to create AI video that fits a specific "style, lens, or cinematic effect." The blog post includes a (pretty cheesy) example of how creators could use the tool -- a video in which a guy's dog grows to monster size. It is both neat and a bit worrying that AI video is so easy to create. YouTube noted that AI-generated creations would feature both SynthID watermarks and clear labels. Hopefully, that will ease confusion from viewers. For the time being, the new AI feature has been launched for users in the US, Canada, Australia, and New Zealand.

Hate AI in Shorts? You might start seeing more of it thanks to YouTube's latest feature

You can use it to generate video backgrounds for your shorts or clips to add to the main video. YouTube is using AI in creative ways to help creators push more content and reach a wider audience. In addition to a tool to automatically dub videos into other languages, the platform introduced a Dream Screen feature last year that lets users generate cool backgrounds for YouTube Shorts with a simple text prompt. Now, it is taking things up a notch with a massive Dream Screen upgrade. With its latest Dream Screen update, YouTube says users can generate standalone video clips for Shorts. The feature utilizes Google DeepMind's newest video generation model, Veo 2, which can generate "state-of-the-art, high-quality videos in a wide range of subjects and styles that better match your vision." YouTube claims that Veo 2 can generate videos faster than before and better understands real-world physics and human movement to deliver more detailed and realistic content. Users can even specify a style, lens, or cinematic effect to match the rest of their footage or give their Shorts a more dramatic look. Creators can use this capability to either add a video background to their Shorts or generate clips to add to the main video. To do the former, navigate to the Green Screen option in the Shorts camera, select Dream Screen, enter a prompt, and select an image. If you want to create a clip instead, you can choose the Add option in the Shorts camera, select Create, input your prompt, select an image, and then choose the desired length. All video clips generated using the feature are watermarked using SynthID and feature clear labels indicating that they are AI-generated creations. The feature is rolling out to users in the US, Canada, Australia, and New Zealand, and YouTube plans to expand to more regions in the future.

You Can Now Create AI-Generated Videos for YouTube Shorts

YouTube said that AI-generated videos will contain SynthID watermark YouTube added a new artificial intelligence (AI) feature for the creators on the platform on Thursday. The video streaming platform announced that Google's Veo 2 AI model is being integrated with the Dream Screen feature. The new video generation model also adds the capability for users to generate standalone AI-generated videos. The company says the tool can be used to add footage that is not easily available or to bring an imagination to reality. Notably, Dream Screen would earlier only allow creators to add video backgrounds using AI. In a blog post, the video-streaming giant highlighted that it is integrating Google DeepMind's latest video generation model Veo 2 with Dream Screen. Using this, users can generate standalone videos to be used in the Shorts. Notably, YouTube Shorts is the short-format vertical scrolling video interface the company launched in 2020. The company said that the feature is aimed at assisting creators on the platform with generating footage that is not easy to find. The tool is a text-to-video generator, so users can simply describe what they are looking for and the Veo 2 AI model will generate it. Veo 2, the latest video generation model, was introduced in December 2024 as the successor of the Veo AI model. At the time of launch, DeepMind stated that Veo 2 made significant improvements in the details and realism of the generated videos. The AI model was also more adept in cinematography and can understand genres, lens types, cinematic effects, and camera movements. To generate videos using Veo 2, users will first have to open the Shorts camera, tap Add to open the media picker, and then tap Create at the top. There, users will see a text field where they can add their prompt. Once done, they can select the style, tap Create video and choose the desired length. YouTube said it will be adding SynthID watermarks to all AI-generated video clips to help mitigate the risk of deepfakes. Notably, the video generation feature is currently only available in Australia, Canada, New Zealand, and the US.

Google Adds AI Video-Generation Model 'Veo 2' to YouTube Shorts

The Dream Screen feature on Shorts gets an upgrade with the integration of Google DeepMind's Veo 2. YouTube Shorts has been integrating AI features for a while now. Previously, it added the ability to generate unique AI backgrounds with just text prompts, followed by an AI-powered video editing app. Last year, it promised to equip content creators with generative AI models, and it has done exactly that. The Dream Screen feature on YouTube Shorts is getting an upgrade with the integration of Google DeepMind's newest video generation model, Veo 2. This means you may not require standalone text-to-video AI models like Goku by ByteDance. To complement this, a new feature was added to generate standalone video clips with the help of Veo 2. Whether you need a specific scene without having any footage on it or just want to create something unique, all you need to do is enter a text prompt to generate a video clip. Along with the feature upgrades, Google has also made under-the-hood improvements to the Dream Screen feature for a faster experience. To generate the clip, you just need to open the media picker from the Shorts camera option and hit 'Create'. Then, you can type in your text prompt, select an image, and choose the desired length. Here's how it works: In a statement, Google said, "Veo 2 understands real-world physics and human movement better, making its output more detailed and realistic. You can even specify a style, lens, or cinematic effect, making Dream Screen an easy and fun way to express yourself." "We use SynthID watermarks and clear labels to indicate that these creations were generated with AI," it added. The AI-powered features are currently available in the US, Canada, Australia, and New Zealand, with plans to expand further.

YouTube Shorts adds Veo 2 so creators can make gen AI videos | TechCrunch

YouTube Shorts is integrating with Google DeepMind's latest video model, Veo 2, allowing creators to generate AI video clips for their posts. Veo 2 is Google's response to Sora, OpenAI's text-to-video generator. YouTube, owned by Google, already allowed creators to generate AI backgrounds for their Shorts with a text prompt through a feature called Dream Screen. But Veo 2 will allow creators to generate standalone video clips to add to any of their Shorts. According to a blog post by YouTube Director of Product Dina Berrada, these AI tools will run faster than before. "Veo 2 understands real-world physics and human movement better, making its output more detailed and realistic," Berrada wrote. "You can even specify a style, lens, or cinematic effect, making Dream Screen an easy and fun way to express yourself." To indicate that these videos were made using AI, they will use DeepMind's SynthID tool, which watermarks and identifies AI-generated content. YouTube also stated that these videos will be clearly marked as being made with AI -- but these labels don't completely mitigate the potential harm of sharing synthetic content that can mislead viewers. To use Veo 2 in YouTube Shorts, creators can open the Shorts camera, select Green Screen, and then navigate to Dream Screen, where they can input a text prompt to generate a video. This feature launches on Thursday for any creator in the U.S., Canada, Australia, or New Zealand. YouTube plans to expand access later.

YouTube Shorts gets AI-powered video generation with Veo 2

On Thursday, YouTube launched a significant update to Dream Screen, its AI-powered tool for generating unique backgrounds in Shorts. Now integrated with Google DeepMind's latest video model, Veo 2, Dream Screen offers more advanced capabilities, taking YouTube Shorts creation to the next level. Dina Berrada, Director of Product for Generative AI Creation at YouTube, announced that Veo 2 introduces a new feature -- standalone video generation. This allows users to create independent video clips and seamlessly integrate them into their Shorts, expanding creative possibilities for content creators. According to Berrada, this update addresses the needs of creators who require specific scenes but do not have the appropriate footage. With Veo 2, users can generate video clips tailored to their vision by simply entering a text prompt. This allows for dynamic and flexible storytelling, opening up new opportunities for creativity. Veo 2 significantly enhances Dream Screen by producing high-quality videos across various subjects and styles. It also improves video rendering speed and realism, offering better physics and more accurate human movement. Additionally, users can customize their creations by selecting specific styles, lenses, or cinematic effects, further personalizing their Shorts. YouTube ensures transparency and accountability by applying SynthID watermarks and clear labels to AI-generated content, helping viewers easily distinguish between human-created and AI-generated media. The new features are currently available in the US, Canada, Australia, and New Zealand. YouTube plans to expand the availability of these capabilities in the near future, with a global rollout expected.

Look out, AI video could soon flood YouTube Shorts

There are some unbelievably great, if abbreviated, films to watch on YouTube Shorts. A lot of them may soon be more literally unbelievable thanks to Google's AI video creation model Veo 2. YouTube has released Veo 2 to the Shorts platform, augmenting YouTube's Dream Screen AI tool and letting you produce AI-fueled flicks based on a text prompt. Dream Screen has been using the original version of Veo to produce video backgrounds out of text prompts for Shorts since last year. Veo 2 ups the ante significantly by also making the characters and objects for the video along with the background. The upgrade also makes Dream Screen faster, better at understanding text prompts, and able to produce much more realistic results. The videos mimic real-world physics, and the characters move as realistically (or cartoonishly) as you might want. You can try out the enhanced Dream Screen by opening the Shorts camera, selecting Green Screen, and typing in what you want to see. You can even add an AI-generated clip to an existing Short by tapping "Add," then "Create," then typing up the prompt. Veo 2 takes over, and within seconds, your giant Pomeranian ballerina is ready to perform. The upgrade to Dream Screen raises many questions and possible concerns. Will AI-generated content flood YouTube Shorts, making it harder to tell what's real and what's not? What will creativity look like when the barriers to high-quality visuals disappear? Will we simply get stuck in a loop of AI-generated influencers making AI-generated content for an audience of AI-powered recommendation algorithms? Google does seem to get that hyper-realistic AI videos made in a few seconds might have some potential pitfalls. That's why YouTube is attaching a SynthID watermark and a label indicating the AI origins of any Dream Screen-produced video. How well these transparency and tracking attempts perform remains to be seen, but at least there's something. The new feature is only coming to the U.S., Canada, Australia, and New Zealand for now, but others are in the pipeline, with more countries on the way. If you're a YouTube content creator, this may be a huge boon, especially if the only thing standing between your video and viral fame is a slightly more perfect shot, better stock footage, or something truly outlandish. If you don't have an idea, you can always toss around ideas with YouTube's Brainstorm with Gemini tool.

YouTube is going to let you put AI-generated video clips in Shorts

The feature is rolling out for everyone in the US, Canada, Australia, and New Zealand starting today, with "plans to expand to more of you in the future," according to YouTube. It's an interesting update given Google's latest Veo model is still in early access and only available to the public via a waitlist. YouTube says the Veo 2 integration allows Dream Screen to generate videos faster, and with more "detailed and realistic" results for real-world physics and human movement. The generated footage will also be tagged with both visual labels and Google's invisible SynthID watermarks identifying that it was created or altered using AI.

Google DeepMind's version of Sora and Adobe Firefly is coming to YouTube Shorts

Summary Google DeepMind's Veo 2 has been integrated into YouTube Shorts' Dream Screen feature, enabling users to generate AI-powered video backgrounds and standalone clips using text prompts. Veo 2 improves the quality, realism, and speed of AI video generation in Shorts, allowing for more detailed and stylized content creation. Users can even specify cinematic effects and lens styles. This new feature is available to users in select regions (US, Canada, Australia, and New Zealand initially), and it doesn't require a Premium subscription. Earlier this week, YouTube shared its roadmap for 2025, complete with four big bets that it's making to ensure that the platform remains the 'epicenter of culture.' As part of the announcement, the streaming giant hinted at improved monetization models, better discovery tools, a suite of creator-focused tools, AI enhancements, and more. YouTube highlighted its AI efforts by spotlighting tools like Dream Screen, Dream Track, auto-dubbing, and its GenAI content inspiration tool that helps creators come up with video ideas that might resonate with their audience. The streaming giant also indicated that it plans to use Machine Learning to better estimate a viewer's actual age, helping it surface more relevant content and prevent inappropriate content from showing up in young users' feeds. Related YouTube's 2025 roadmap has four big bets, and only one focuses on AI Machine Learning to make YouTube safer for everyone Posts 1 Now, fulfilling a part of its bet, YouTube has rolled out a new AI feature -- one that brings DeepMind's Veo 2 over to Shorts. The new feature joins Shorts' virtual AI-generated green screen tool called Dream Screen, which allows users to seamlessly generate backgrounds, or standalone shorts, only by using conversational prompts. Now, with Veo 2, in addition to text-to-image background, users can also create short text-to-video clips to use as video backgrounds. And, best of all, the tool is also available to non-Premium users, albeit only in the US, Canada, and New Zealand, at least for now. It's worth noting that all AI-generated content will be labeled using SynthID watermarks and clear labels. Veo 2 understands real-world physics and human movement better, making its output more detailed and realistic. You can even specify a style, lens, or cinematic effect, making Dream Screen an easy and fun way to express yourself. Related YouTube wants you to use an AI-generated green screen in its latest Shorts test The tool is only available to a small number of creators, but we have a workaround Posts Here's how you can begin experimenting with Veo 2 on YouTube today Open YouTube and tap the plus button at the bottom. Tap the Dream Screen icon on the right (it looks like a human silhouette) Tap on the pink-purple gradient Gemini box at the bottom. Describe what you would like to create and choose a style. These include Vintage, Anime, Clay, Digital, and more. Chose one of the four image templates to convert into video and tap Create Video. Follow along the screenshots below for visual cues Close

Here's How You Can Create AI Videos in YouTube Shorts, Thanks to Google Veo

Katelyn is a writer with CNET covering social media, AI and online services. She graduated from the University of North Carolina at Chapel Hill with a degree in media and journalism. You can often find her with a novel and an iced coffee during her time off. Creators looking to play around with AI video tech are in for a Valentine's Day treat. YouTube Shorts has new AI video capabilities, thanks to Google's AI video model, Veo 2, being integrated into the video platform. In September at its Made on YouTube event, YouTube first said Veo would add more firepower to its existing Dream Screen tools. This week, the company announced the new tools are now available for folks in the US, Canada, Australia and New Zealand. The new features let you create short, silent AI clips to include in your video, and you can record yourself in front of an AI-generated background image or video. Google's AI policy when you use AI on Shorts says YouTube "collects your prompts, related content, and your feedback" to develop YouTube's services. All AI videos are marked with Google's SynthID, an invisible watermark that designates that it was created by generative AI. AI has been slowly but steadily creeping into our daily social media experiences. Professional creators and businesses have seen AI-powered updates behind the scenes for some time, and now regular users are seeing more of the tech. When Meta integrated its AI across its social platforms, users quickly noticed (and complained about) how prevalent it was on Instagram, Facebook, Messenger and WhatsApp. TikTok, though busy dealing with other challenges, has also been rolling out AI avatars and other features for creators. As AI video tech becomes more accessible and refined, it's likely to appear a lot more frequently in our video-first social media environments. YouTube integrating Google Veo is just one of the first examples of this, and creators' and users' responses to it will be telling about the future of AI-video on social. If you want to try out the new YouTube AI video tools for yourself, here's how. To begin, open the YouTube app and tap the plus button to start creating a new Short. Here's how you can create an AI-generated background in a new Shorts video. Once you're finished, tap the check mark and upload your video like normal. The AI-generated background videos aren't super long, so you might get cut off while you're recording. In my testing, I found it was easiest to break down my script into shorter clips and record several of them instead of recording in one take and risk getting cut off. To create an entirely new clip with AI, tap the plus sign to begin creating a new video and follow these steps.

YouTube Shorts gets AI upgrade with Google DeepMind's Veo 2: Now available in the US and more countries

YouTube's Dream Screen once only offered AI-generated backgrounds for Shorts, but with Veo 2, creators can now generate full-fledged video clips, adding a new level of creativity and realism to their content.YouTube Shorts is collaborating with Google DeepMind's new video model, Veo 2, to help creators generate AI-powered video features for their Shorts. YouTube, owned by Google, has revealed that the model will be integrated into YouTube's Dream Screen, an experimental text-prompt feature. It will also allow users to create standalone video clips to add to their Shorts. Launched in September 2023, Dream Screen allows creators to generate images and videos for Shorts using text descriptions. But until now, these could only be used as virtual green screen backgrounds. To generate clips, Dream Screen users must open the Shorts camera, select the media picker, and tap the "Create" option at the top. They can then type a text prompt for the content they want and choose their preferred style, lens, cinematic effect and video length. The feature is now available in the US, Canada, Australia and New Zealand, with plans to expand to more countries in the future, according to YouTube. Also Read : Apple exploring humanoid and AI-powered robots to revolutionise smart homes, says analyst This update is particularly intriguing, as Google's new Veo model is still in early access and is only available to the public through a waitlist. YouTube claims that Veo 2 integration enables Dream Screen to create videos more quickly, delivering results that are more "detailed and realistic," particularly in terms of real-world physics and human movement. Generated videos will be marked with visual labels and will also feature Google's hidden SynthID watermarks to indicate they were created or modified using AI. Also Read : Kanye West returns to X three days after controversial posts -- Offensive content removed 1. What is Veo 2? Veo 2 is an advanced video generation model developed by Google DeepMind. It is designed to create highly realistic and detailed video content based on text descriptions. The model can generate video clips with accurate depictions of physics and human movement, making it ideal for applications such as YouTube's Dream Screen feature. 2. How is it different from the previous model? Compared to its predecessor, Veo 2 offers faster video generation, improved accuracy and more lifelike results, particularly in the way real-world physics and human actions are portrayed. It also provides more detailed and realistic video effects, addressing some of the limitations of the earlier model.

I Tried YouTube's AI Video Generator, and I'm Scared

This week Google added Veo 2 to YouTube Shorts and -- wow, this article is already gibberish. Let me back up. YouTube Shorts is YouTube's TikTok competitor. Veo2 is Google's AI tool for generating videos. As of this week, you can use that AI to make clips for YouTube Shorts. This innovative and revolutionary change will forever alter the vertical videos most people scroll past so they can find the thing they actually wanted to watch. According to a confusingly worded announcement post, you can now create clips to add to your Shorts simply by typing a sentence. Anyone can create a Short right in the YouTube mobile app either by combining footage already saved to their phone or by recording directly. Now there's a third option: generating horrifying footage using AI. "Need a specific scene but don't have the right footage?" the post asks. "Simply use a text prompt to generate a video clip that fits perfectly into your narrative, or create a whole new world of content." I can't think of any place less appealing than a "whole new world of content." The very phrase conjures up mental images of a digital purgatory, an endless void of placeholder content where the unworthy are sent to after a life lived passively consuming. So naturally I wanted to try this out. I fired up the YouTube application on my phone found that the feature is a little buried. To generate a clip, you need to click the Add button in the bottom-left corner and then -- instead of clicking one of the photos or videos on your phone -- tap the Create button at the top of the screen. After that, you can type your prompt and tap Create. You'll see four thumbnails -- tap one and the AI will generate a clip for you. This will be added to your timeline, meaning you can edit it into the rest of the Short however you like. I played around with this a few times, mostly with cats eating planets (I wanted to see it). The results were not particularly convincing from a feline or astronomical perspective -- I'd say it resembled the feeling of a fading memory of a dream more than actual video footage -- but I certainly ended up with a video that included the elements I described. There's one more way you can use these videos -- as an animated background for your Shorts. To do this, tap the down arrow on the toolbar to expand it, tap the Green Screen option, then tap the AI sparkle instead of one of your photos. As before, you will be given four thumbnails to choose from -- the one you choose will become a video. You can then record yourself talking as the video loops in the background. Honestly, this seems like a decent use for AI -- a quick and disposable visual flare for a video that is also quick and disposable. For what's it's worth, we have an article about how to hide YouTube Shorts from your feed, as does seemingly every other website on the internet. I wonder why that is.