Sora Turbo: Open AI Launches New AI Video Generator

OpenAI announced on Monday the launch of its new video generation model, Sora Turbo, moving the technology out of its research phase and into public deployment.

This tool allows ChatGPT Plus and Pro subscribers to create realistic videos from text prompts. The updated Sora Turbo model significantly outperforms its February preview version in speed and functionality.

Users can generate videos in resolutions up to 1080p, with durations of up to 20 seconds. The tool supports various aspect ratios, including widescreen, vertical, and square formats.

Sora’s new interface includes features like a storyboard tool for frame-by-frame input specifications, options to extend, remix, or blend existing assets with newly generated content, and feeds showcasing featured and recent community creations.

Sora’s release follows a recent controversy where a group of artists, claiming involvement in OpenAI’s alpha testing program, leaked the product in protest, alleging it was used for unpaid labour.

day 3 -✨sora

rolling out starting today at https://t.co/9chu6jWk9w. have fun! pic.twitter.com/m2OrZef7Zy

— shyamal (@shyamalanadkat) December 9, 2024

Sora is integrated into ChatGPT Plus accounts at no additional cost, allowing up to 50 videos per month at 480p or fewer at 720p. Pro subscribers gain higher resolutions, longer durations, and tenfold usage limits. Tailored pricing plans for diverse user needs are expected early next year.

Recognizing the challenges of realistic AI-generated video, Sora’s developers have implemented safeguards to ensure responsible use.

All videos include metadata compliant with the Coalition for Content Provenance and Authenticity (C2PA) standard, providing transparency and traceable origins. Watermarks are enabled by default, and an internal search tool helps verify Sora-generated content.

At launch, the system restricts the generation of harmful material, such as child exploitation content and sexual deepfakes. Uploads featuring real people will initially be limited to refine safety protocols.

Despite its advancements, Sora Turbo faces limitations, such as challenges with realistic physics and complex actions in longer videos.

Non-subscribers can still browse the feed of AI-generated videos created by others using Sora. While the model launches today in the U.S. and several other countries, OpenAI CEO Sam Altman stated that availability in “most of Europe and the UK” may take some time, as reported by The Verge.

 

View this post on Instagram

 

A post shared by TechCrunch (@techcrunch)