Last Updated on
Over the past few years, the advancements in AI technology have created a space for impressive, new software to take center stage, and we expect OpenAI’s upcoming Sora tool to be no different. OpenAI has recently unveiled a new video-creating tool, that creates a realistic generated video from text and photo prompts.
This new technology is set to impress the public and potentially change the filmmaking world as we know it, but what is Sora? And, how does it work? In this article, we answer these questions and more; covering Sora’s capabilities and features.
If you’re keen to find out more about AI-generated video software, then check out our guide on Google Lumiere or our best AI video generators.
What is OpenAI Sora?
Sora is an OpenAI text-to-video model that generates videos with complex scenes that are up to a minute long. Sora creates these videos from the user’s descriptive captions and still image prompts. OpenAI has specified that the main goal of Sora is for real-world use; helping people solve problems that require real-world interaction by training AI to understand the physical world in motion.
Essential AI Tools
Content Guardian – AI Content Checker – One-click, Eight Checks
Best Deals
Jasper AI
Best Deals
WordAI
Best Deals
Copy.ai
Best Deals
Writesonic
Best Deals
What can OpenAI Sora do?
OpenAI Sora can create entire videos with detailed scenes that maintain visual quality and visual style, just from a user’s prompt. These videos can include multiple characters, specific types of motion, vibrant emotions, depth of field, and accurate details of the subject and background. Sora was created with a deep understanding of language, so it can efficiently produce accurate interpretations of prompts; understanding what the user’s prompt is detailing and how it exists in the real world. Another useful feature of the model is its ability to take existing videos and extend them or fill them with missing frames
With the benefits of creating such a tool, arise issues surrounding hateful content and what the platform should and shouldn’t do. During its ongoing testing and extensive research, OpenAI has said they’re working with red teamers (domain experts in areas like misinformation and bias) to create safe AI systems. Mainly focusing on detecting prompts that could generate harm, like sexual content, celebrity likeness, extreme violence, the IP of others, and hateful imagery.
Can I use OpenAI Sora?
OpenAI Sora is not yet available for public use and OpenAI has not specified a date on which it will be released to the public. At the minute it is only available for select creators and security experts who are in the process of testing weaknesses and providing feedback for research progress.
The current model is not ready for public use because of its apparent weaknesses in critical areas. As stated by OpenAI, Sora currently struggles to represent the physics of a complex scene. For example, a character in a generated video might take a bite out of a cookie but the cookie will be left with no bite mark after. Additionally, it has been noted that the generator might confuse the spatial details of a prompt (mixing up left and right), so the goal of training models is to fine-tune and solve these issues.
What are the limitations of OpenAI Sora?
Like many AI applications, OpenAI Sora cannot be used to generate videos depicting violence, adult content, or that of real people or in the style of named artists. Upon its announcements, immediate questions have also been raised on copyright infringement by filmmakers and creative professionals, as well as the program’s safety and veracity – particularly in the spread of misinformation. Despite researcher Bill Peebles claiming that “training data is from the content we’ve (OpenAI Sora) licensed and also publicly available content”, several lawsuits regarding AI’s use of “publicly available” copyrighted content is currently underway.
Wrapping up
OpenAI Sora provides a real sense of what AI capabilities can do, and the current model, despite its apparent issues, is still an impressive step for OpenAI platforms. After the model is out of the red-team stage and available to the public it is sure to make strides in the world of AI-generated video tools. Stay tuned as we keep up-to-date on everything being announced about this platform.