Text, images and now videos — OpenAI's latest offering Sora is indeed an advancement in AI technology, specifically focusing on generating videos from text inputs. It marks a significant step forward in the field of AI, following the development of tools like ChatGPT and DALL-E.
Sora isn't the first of its kind - Google, Meta, and the startup Runway ML have dabbled in similar technologies, but OpenAI's Sora is turning heads with its astonishing video quality.
here is sora, our video generation model:https://t.co/CDr4DdCrh1
today we are starting red-teaming and offering access to a limited number of creators.@_tim_brooks @billpeeb @model_mechanic are really incredible; amazing work by them and the team.remarkable moment.— Sam Altman (@sama) February 15, 2024
The buzz around Sora reached a fever pitch when CEO Sam Altman invited social media users to throw out written prompts. The responses were eclectic, ranging from a grandmother influencer hosting a rustic cooking session for homemade gnocchi to a bicycle race on the ocean with different animals as athletes.
Yet, within moments, Sora transformed these words into stunningly realistic videos that left onlookers spellbound.
https://t.co/qbj02M4ng8 pic.twitter.com/EvngqF2ZIX
— Sam Altman (@sama) February 15, 2024
Here's a breakdown of some key points about Sora:
Functionality: Sora is a text-to-video diffusion model, capable of generating videos based on textual descriptions. Users can input text prompts, and Sora will create corresponding video content.
Capabilities: Sora is capable of producing realistic and complex videos of up to one minute in length. It can create detailed scenes with multiple characters, providing a wide range of possibilities for creative expression.
Safety measures: OpenAI is taking precautions to ensure the responsible use of Sora. This includes engaging in red-teaming exercises to identify potential risks and harms associated with the model. Additionally, OpenAI is developing tools to label Sora-generated videos and applying safety methods similar to those used with DALL-E to reject inappropriate or harmful text prompts.
Engagement with stakeholders: OpenAI is actively engaging with policymakers, educators, and artists to understand their concerns and identify positive use cases for Sora. This reflects a commitment to fostering dialogue and collaboration around the ethical and responsible deployment of AI technologies, OpenAI said.
Comparison to other models: While other video-generating models exist, Sora stands out for its ability to produce realistic and complex videos. Meta has a tool for creating short video clips, and Google is working on its text-to-video model, but Sora is purportedly capable of more advanced video generation.
However, despite its impressive capabilities, Sora is not yet available to the public, and OpenAI has disclosed limited information about its development process. Concerns have been raised about the sources of imagery and video used to train Sora, especially considering OpenAI's past legal issues related to the use of copyrighted works.
First Published: Feb 16, 2024 9:56 AM IST
Check out our in-depth Market Coverage, Business News & get real-time Stock Market Updates on CNBC-TV18. Also, Watch our channels CNBC-TV18, CNBC Awaaz and CNBC Bajar Live on-the-go!
Lok Sabha Election 2024: What rural Delhi wants
May 16, 2024 10:10 PM
Over 50 onion farmers detained in Nashik ahead of PM Modi's visit
May 16, 2024 11:14 AM
Why Google CEO is cautiously optimistic about the election year
May 16, 2024 9:51 AM
Mark Mobius reveals how markets will react if NDA wins 400+ Lok Sabha seats
May 15, 2024 8:09 PM