The company behind the ChatGPT AI tool, OpenAI, has introduced a new AI Text-To-Video Generator called Sora. They made this known in a blog post they made on Thursday.
This new AI tool creates realistic videos based on text prompts. According to the blog post, Sora has “a deep understanding of language” and can generate “compelling characters that express vibrant emotions.”
They also made it known that “Sora is able to generate complex scenes with multiple characters, specific types of motion, and accurate details of the subject and background,
The model understands not only what the user has asked for in the prompt, but also how those things exist in the physical world.”
They explained “ We are teaching AI to understand and Simulate the physical world in motion, with the goal of training models that help people solve problems that require real-world interaction.”
On his X page, the CEO of OpenAI Sam Altman, invited X users to try out the AI tool by suggesting prompts for the tool. Some people gave it a try and they got results of realistic videos of golden retrievers podcasting on a mountain top, a grandmother making gnocchi and so on. After this trial by some X users, the internet has been blowing up as people have been mesmerized by the results of this new AI tool. One X user said “It’s been two hours and my brain still can’t process these generated OpenAI Sora videos.”
Nevertheless this newly launched tool also raised some concerns amongst various people because of the potential danger and manipulations associated with this type of tool most especially in a year where there will be closely watched elections worldwide. A good example is the upcoming US presidential election which will be coming up later this year in November.
However, Open AI has put the minds of people at rest, Sora is not yet open to the general public, as the company is still at work on how to improve the safe use of the tool. For example how to make the tool reject harmful inputs, sexual contents, hateful images and so on. In order to achieve this goal, OpenAI is working with industry experts. They said, “Despite extensive research and testing, we cannot predict all of the beneficial ways people will use our technology, nor all the ways people will abuse it. That’s why we believe that learning from real-world use is a critical component of creating and releasing increasingly safe AI systems over time”.
OpenAI also made it known that Sora has its own weaknesses such as difficulty with continuity and distinguishing left from right.“For example, a person might take a bite out of a cookie, afterwards, the cookie may not have a bite mark.”
OpenAI is not the first company to launch a Text-To-Video AI technology. Other companies such as Meta and Google have also launched their own models but their models have not produced the realistic results that OpenAI’s Sora has.