What’s New in Veo 3.1, Google’s Latest AI-Powered Video Platform
Did you know you can configure Google to filter out junk? Follow these steps to improve your search results, including adding my work on Lifehacker as a preferred source .
OpenAI’s new app, Sora, has become a hot topic in the discussion of hyperrealistic AI problems in recent weeks. Sora allows users to easily create short videos that most people find realistic, including videos that demonstrate similarity to real people .
But before Sora went down, it was Google that was sowing fear about these realistic AI-powered videos. With the release of Veo 3, Google launched an AI model that not only created realistic videos but also generated realistic audio synchronized with the action. Sound effects, environments, and even dialogue could be generated in parallel with the video itself, fully realizing the effect with a single simple command.
Veo 3.1
Now Google is back with an update to Veo, appropriately named Veo 3.1, which the company announced in a blog post on Wednesday . It’s not necessarily a complete overhaul or a revolutionary new video model. Instead, Veo 3.1 builds on Veo 3, adding “richer audio” and “heightened realism,” which Google says generates “lifelike” textures. The new model also reportedly supports new narrative management tools, which align with new updates to Flow, Google’s AI-powered video editor. Flow users now have more granular control when editing and can add audio to existing features like “Ingredients in Video,” “Frames in Video,” and “Expand.”
What does this mean in practice? According to Google, the Ingredients to Video with Veo 3.1 feature allows users to add reference images to scenes, such as images of specific people, clothing items, or environments. The new Flow editor then allows these elements to be inserted into the finished product, as shown in the demo video below:
With this new feature, Flow now lets you add new elements to an existing scene. Using the “Insert” feature, you can instruct Veo 3.1 to add new characters, details, lighting effects, and more to a clip. Google says it’s also working on a reverse solution: users will be able to remove any unwanted elements from the generated image.
Google also offers users a new way to specify how scenes are generated: “First and Last Frame.” Users can select keyframes for the beginning and end of a scene. Flow with Veo 3.1 will fill the gap and generate a scene that begins and ends based on these images.
You can now create longer videos than in previous versions of Flow. The new “Extend” feature allows you to either continue the current clip or jump to the next scene. Google says this feature is most useful for creating longer introductory shots. According to the company, Extend allows you to create videos longer than a minute.
Veo 3.1 is available to Gemini app users, as well as Vertex AI users with a Google AI Pro subscription . Developers can access it through the Gemini API. Google reports that the “Video Ingredients,” “First and Last Frame,” and “Expand” features will be available in the Gemini API, but the “Add Object” and “Remove Object” features are not yet available. The “Expand” feature is also not yet available in the Vertex AI API.
Is this really good?
Google sees all these advances as a boon for creators and their work, but I’m highly skeptical. I could see Veo 3.1 and Flow as a good tool for visualizing shots before filming or animating them (i.e., a storyboard tool), or even as a way for aspiring filmmakers to master editing by seeing their ideas more clearly. Overall, however, I don’t think AI-generated content is the future, or at least not the future most of us want. Sure, some of these AI-generated videos have humor or novelty, but I’m willing to bet that most people who enjoy them do so ironically or exclusively on social media.
The idea of replacing human directors and actors with generations of artificial intelligence seems absurd, especially considering it exposes us all to the risk of misinformation. Is it really so important for companies like Google and OpenAI to simplify the creation of hyper-realistic, fully rendered scenes when these videos are so easily used to deceive the masses? Perhaps this is the ravings of someone resistant to change, but I don’t think most of us would like to see our favorite shows and movies, made with soul and emotion, replaced by realistic-looking humans playing voiceless and robotic roles.