OpenAI is currently on an innovation spree and has launched a new text-to-video AI tool called Sora that will be able to create minute-long complex videos with multiple characters and elements.
The announcement came in on Thursday 17 February and while the tool isn’t available publicly yet, a few examples have been shared online which has sparked a debate online.
The examples are available on Sam Altman’s X account. One of the examples shows a command “Two golden retrievers podcasting on top of a mountain” and a video showing just that— two golden retrievers with a black headphone each, sitting on a beautiful mountain top on a red and white sheet, with a mic in between. The video is so realistic that it’s hard to tell if it’s AI-generated.
The video quickly went viral, garnering over 56k+ likes so far.
Sam has also asked his followers to submit their own prompt ideas for the tool. Sora can do a lot more than just text-to-video animation. For instance, if you already have a video, you can use Sora to extend it and add some extra frames. This will make video restoration or editing very easy.
Similarly, if you have a photo, you can add it to Sora and it’ll animate the picture for you. From one still photograph, you’ll get a realistic video where every small detail is accurate. This can be helpful in recreating your favorite memories, especially the ones that can’t be recreated in real life.
Why Are Some People Concerned About The Tool?
While a certain group of netizens were overjoyed at the idea of being able to swiftly bring their ideas to life, others raised concerns over how dangerous this tool could be.
One of the concerns was that content creation is a growing industry that has created a lot of job opportunities. But with AI automating it, it could lead to a number of job cuts. At a time when layoffs are already a big issue, Sora could make things a lot worse.
For instance, Oren Etzioni, founder of TruMedia.org said “Generative AI tools are evolving so rapidly, and we have a social network — which leads to an Achilles heel in our democracy and it couldn’t have happened at a worse time,”
OpenAI shares these sentiments and hence has decided not to release it to the public yet, at least not until researchers mark it safe.
A recent blog shared by OpenAI states that the tool will only be available to a few selected creators for now and access will be limited in the initial phase.
Then, before it completely integrates with OpenAI’s other products, a team of experts will assess the app for its safety.
Also, to address the growing trend of AI misuse, OpenAI is also working on a tool that’ll help identify the videos generated by Sora. Plus, they have promised to train the model to reject violent and sexually explicit prompts to minimize its misuse.
Share 
Krishi Chowdhary Journalist 
Question & Answers (0)