Google's new anything-to-anything AI model is wild

Omni Flash is a generative model family from Google that creates video from inputs such as photo, video, and text. Omni Flash is available in Google’s Flow platform for AI video generation and editing. Users can upload a video and provide a text prompt to guide the AI’s output. Omni is claimed to use more real-world knowledge and maintain character consistency better than the earlier Veo model. Testing with a deepfaked stuffed animal showed mixed results: some clips matched prompts more closely and kept characters consistent, but some still included sudden orientation changes and other AI jump-scare artifacts. Overall, the tools require little effort and are rapidly improving.

"Last year I deepfaked my kid's stuffed animal to make it look like his plush deer was on vacation. It was an experiment to see if I could re-create the events depicted in a Gemini ad Google was running, and I never showed the videos of Buddy the deer on his adventures to my four-year-old. But it was a revealing exercise that made me think a lot about the difference between some harmless fun with generative AI and full-on slop."

"Omni is a new family of generative models that will allegedly one day be able to turn any kind of input - photo, video, text - into anything else. But for starters, it's just creating video. Omni Flash is the first of these models Google has released, now available in the company's AI video generation and editing platform, Flow. You can still use the previous model, Veo, if you want, but Omni improves on Veo in a few ways."

"With Omni, you can upload a video and use that along with a text prompt as the starting point for your AI-generated creation. Google also claims Omni incorporates more real-world knowledge when producing videos and can do a better job of keeping characters consistent throughout a video as a result. There was only one way to really know if those claims are true: I brought back AI Buddy to pack his little AI-generated bags for another adventure."

"The results are such a mixed bag they're baffling. Some were very good - much more consistent and true to my prompt than when I was testing out Veo five months ago. But even the best clips Omni cooked up for me still have certain AI jump scares, like when Buddy suddenly switches orientation while he's skydiving."

Read at The Verge