Open Sora is the latest gem in Sam Altman's OpenAI empire. Just when skepticism loomed over the company's future amid last year's internal tussle between positionings, the non-profit AI firm came in with a bang in the digital sphere with the launch of Sora, a text-to-video model.
What Is Sora?
Quite similar to Dall-E, which uses text prompts to create an image, Sora uses text prompts to create short videos.
Sora can produce videos lasting up to one minute, just by using a simple prompt. "Sora can generate videos up to a minute long while maintaining visual quality and adherence to the user’s prompt," the site blog read.
The AI model can animate a static image into a video presentation.
"The model is able to take an existing still image and generate a video from it, animating the image’s contents with accuracy and attention to small detail," it further stated. Sora can also extend existing videos or fill in gaps by generating missing frames.
Here is a quick guide on how it works-
Give a written prompt around what you are looking for
The AI model will create a video (max. 1 minute)
For creating longer videos, you can add to the prompts and synchronize them with frame count
Is It available To The Public?
Currently, the text-to-video model is accessible exclusively to policymakers, educators, and artists. This restrictive access aims to ensure the security and safety of the AI model while also gathering valuable feedback on enhancing its capabilities to help creative professionals.
The newly launched AI software will be accessible for red teaming purposes as well, which will assist the company in identifying any weaknesses in its development that could lead to misinformation, bias, and hateful content.
For now, Sora is unavailable for public use, and OpenAI has provided limited details regarding the exact date of its full release to the public.
The Evergreen Debate On AI
Following the launch of Sora, quite evidently, netizens couldn't keep their calm. The evergreen debate of 'AI being more harm than help' resurfaced, as it often does whenever AI comes under the mainstream spotlight.
While the software promises to be a game-changer for many professionals and students alike, concerns about deepfakes loom large. Instances like the creation of a fake video featuring Rashmika Mandanna and the manipulation of PM Modi's voice toned into Bollywood songs raise valid worries about privacy violations and the potential for misinformation to spread.
Although OpenAI has made itself clear that it will be using text and image classifiers to automatically reject those prompts that go against ethical guidelines, considering the technology is still in the crawling stage, things are easier said than done.
When it comes to AI, making a conclusive statement is hard, as the technology is still developing. So every time it takes a step forward, more than optimism and pessimism, it's the mix of the two that dominates the minds. Innovation is a sure thing but at what cost remains uncalculated.
