{"id":53163,"date":"2026-05-27T21:27:08","date_gmt":"2026-05-27T21:27:08","guid":{"rendered":"https:\/\/www.europesays.com\/ai\/53163\/"},"modified":"2026-05-27T21:27:08","modified_gmt":"2026-05-27T21:27:08","slug":"googles-new-gemini-omni-ai-video-model-can-do-crazy-things","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ai\/53163\/","title":{"rendered":"Google’s New Gemini Omni AI Video Model Can Do Crazy Things"},"content":{"rendered":"

Google\u2019s new Gemini Omni artificial intelligence (AI) model can do some wild things. The model\u2019s key promise<\/a> is to create anything from, well, anything. <\/p>\n

Google says its new Gemini Omni model can \u201ccreate anything from any input,\u201d including audio, video, photos, and text. The model starts with video generation, which users can then edit via conversational text with Gemini. This first model, Gemini Omni Flash, is launching now in the Gemini app, Google Flow, and YouTube Shorts. <\/p>\n

As Google explains, editing AI-generated video using text is straightforward. The model also promises to keep things consistent after editing, including characters, and Omni can remember what was visible in previous scenes. <\/p>\n

Prompt: Make the sculpture out of bubbles. <\/p>\n

The company even promises that Gemini Omni can use its \u201cintuitive understanding of physics,\u201d effectively \u201cbridging the gap from photorealism to meaningful storytelling.\u201d <\/p>\n

Prompt: A marble rolling fast on a chain reaction style track, continuous smooth shot. <\/p>\n

Users have already achieved impressive results with Gemini Omni. For example, ex-Google product manager Bilawal Sidhu gave Gemini Omni a photo with a sketched drone path on it and had the AI generate drone POV footage. <\/p>\n

Gave google omni a sketched camera path and asked it to generate drone POV footage. pic.twitter.com\/cQZFMtOkEi<\/a><\/p>\n

\u2014 Bilawal Sidhu (@bilawalsidhu) May 26, 2026<\/a><\/p>\n

The Verge\u2018s Allison Johnson calls Omni \u201cwild,\u201d and had the AI bring her child\u2019s stuffed animal, Buddy, to life<\/a>. Buddy went on exciting AI adventures, including white-water rafting and snowboarding.<\/p>\n

\u201cThe results are such a mixed bag they\u2019re baffling. Some were very good \u2014 much more consistent and true to my prompt than when I was testing out Veo five months ago,\u201d Johnson writes. \u201cBut even the best clips Omni cooked up for me still have certain AI jump scares, like when Buddy suddenly switches orientation while he\u2019s skydiving.\u201d <\/p>\n

Prompt: turn this into realistic footage, using the drawing only as a guide for movement, do not show the drawing in the final video <\/p>\n

As Johnson tested, Omni\u2019s biggest claim to fame, being able to combine a wide variety of input media with AI-generated video, veers from technologically impressive to potentially hazardous. One of her deepfakes even convinced her husband, \u201ca man who has looked at me in real life basically every single day for the last decade.\u201d <\/p>\n

Whether this is neat or terrifying depends on who is asked. <\/p>\n

\u201cI can\u2019t be the only one to think, that this just has no reason to exist,\u201d writes near_photography on Threads in response to Johnson\u2019s post above. \u201cThere is no net benefit to society from this capability.\u201d <\/p>\n

Prompt: Apply the pose and motion from input video to provided character from this image. Apply style from image reference to the new video <\/p>\n

As Google notes, all videos generated using Omni include its \u201cimperceptible SynthID digital watermark,\u201d which makes it easy for users to confirm if something was made with Google\u2019s AI inside Gemini, Gemini in Chrome, and Google Search. But what if someone isn\u2019t using those platforms? <\/p>\n

Google is bringing this technology directly into YouTube Shorts and YouTube Create, for example, and it\u2019s impossible to predict what people will do with it there. <\/p>\n

Image credits: Google <\/p>\n