Sora: OpenAI has launched a tool to instantly create video from text
Last Thursday, OpenAI unveiled a tool called Sora that can create videos from text prompts.
The new model, nicknamed Sora after the Japanese word for “sky,” can also produce photorealistic shots up to one minute long that adhere to the user’s instructions regarding subject and style. According to the company’s blog, the model is also capable of creating a video based on a still image.
“We teach AI to understand and simulate the moving physical world, with the goal of training models that help people solve problems that require real-world interaction,” the blog post reads.
One of the embedded videos was among several initial examples from the company based on the following text: “Movie trailer featuring the adventures of a 30-year-old astronaut wearing a red wool knit motorcycle helmet, blue sky, and salt desert, cinematic style, shot on… 35mm film, vivid colours.
The company also announced that it has opened up access to Sora to a small number of researchers and video creators. While experts will “red team” the product — testing it to see if it can circumvent OpenAI’s terms of service, which prohibit “extreme violence, sexual content, hateful images, likenesses of celebrities, or the intellectual property of others,” according to a blog post. Company. The company only allows limited access to researchers, visual artists and filmmakers, though CEO Sam Altman responded to user calls on Twitter after the announcement with videos he said were made by Sora. Videos carry a watermark to show they are produced by artificial intelligence.
The company also debuted still image generator Dall-E in 2021 and AI-generated chatbot ChatGPT in November 2022, which quickly amassed 100 million users. Other AI companies have rolled out video creation tools for the first time, though these models have only been able to produce a few seconds of footage that is often unrelated to their claims. Google and Meta have said they are developing generative video tools, though they have not released them publicly. On Wednesday, it announced an experiment to add deeper memory to ChatGPT so it can remember more of its users’ conversations.
OpenAI did not disclose how much footage was used to train Sora or where the training videos may have originated, other than telling the New York Times that the corpus contained videos that were both publicly available and licensed from copyright owners. The company has been sued multiple times for alleged copyright infringement in the training of its generative AI tools, which digest gargantuan amounts of material scraped from the internet and imitate the images or text contained in those datasets.
Conclusion
In conclusion, it seems that OpenAI has taken a big step into the future with the launch of the Sora tool, which is capable of converting texts into high-quality videos. As AI research continues and develops, innovators and researchers will be able to express their ideas in new and unique ways, opening new horizons for innovation. Despite the ethical and legal challenges associated with copyright and use of content, the focus remains on the positive impact this progress can have on society. Concerned institutions must keep pace with these developments and ensure that they are used in ways that benefit everyone in a responsible and creative manner.
