Skip to content

  • Home
  • Top News
  • World
  • Economy
  • Science
  • Tech
  • Sport
  • Entertainment
  • Contact Form

REPRESENT: Former Snap AI chief releases AI-powered video generator

REPRESENT: Former Snap AI chief releases AI-powered video generator

Theodore Meeks, April 8, 2024

When you take a selfie and presto, you can learn to walk a longboard quickly. This sounds more impressive than it is. Because actually it is a video created with AI Diffuse. With the app, all you need is a selfie and you can get such animated poses. Fun dances, other sports and more are possible.

advertisement

Diffuse is available as an app for iOS, but not in Germany yet. However, the development is exciting. Not least because OpenAI, for example, is still keeping its video AI Sora under wraps. The company offering Diffuse is called Higgsfield and was founded by Alex Mashrabov, the former head of AI at Snap. Snap integrated its own AI-powered chatbot early on, and AI is also being used for filters and augmented reality effects on Snapchat.

Video creator is popular as a social networking tool

The new application is based on a custom model for converting text to video. In addition to the selfie option, you can also create videos with fictional content via prompting. However, it's clearly about the fun and social factor, not the professional videos. An example on the site is to send your friend a video of yourself performing cheerful movements that you may not be able to do yourself in front of the camera. However, the videos on Higgsfield's homepage are also impressive: very realistic-looking women smiling at the camera, cars driving on sand, and a man running with the head of a rabbit.

“We are a grassroots model company that wants to democratize social media creation,” she says. There is a waiting list for those who want to test out a particularly “powerful and realistic mode.” Nothing further is known about the model or the planned paid versions.

Sora will also be made available to paid audiences this year, CTO Mira Murati recently revealed in an interview. However, OpenAI sees the high potential for AI misuse of video, and says it is still keeping the matter under wraps. Some video artists have already been allowed to test it. Videos still contain the usual artifacts and inaccuracies produced by AI image generators and now video generators. You can also see this very clearly in the examples from Diffuse – cars crashing into each other. If you want to post a funny video or send it to a friend, it's probably less annoying than if the filmmakers had to struggle with it.


(Emo)

To the home page

Theodore Meeks

Lifelong foodaholic. Professional twitter expert. Organizer. Award-winning internet geek. Coffee advocate.

Tech

Post navigation

Previous post
Next post

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Navigate

  • Home
  • Top News
  • World
  • Economy
  • Science
  • Tech
  • Sport
  • Entertainment
  • Contact Form

Pages

  • About Us
  • DMCA
  • Contact Form
  • Privacy Policy
  • Editorial Policy

Pages

  • About Us
  • Contact Form
  • DMCA
  • Editorial Policy
  • Privacy Policy

STAY UPTODATE

Get the Latest News With Aviationanalysis.net

OFFICE

X. Herald Inc.
114 5th Ave New York,
NY 10011, United States

QUERIES?

Do you have any queries? Feel free to contact us via our Contact Form

Visit Our Office

X. Herald Inc.
114 5th Ave New York,
NY 10011, United States

©2025 | WordPress Theme by SuperbThemes
  • Home
  • Top News
  • World
  • Economy
  • Science
  • Tech
  • Sport
  • Entertainment
  • Contact Form