New AI Software Helps Create Realistic Videos Using Just Text Descriptions

OpenAI, the company that owns ChatGPT, is taking its artificial intelligence technology to the video category. The company has just introduced Sora, its next-generation AI model. Sora works similarly to OpenAI's existing image-generating AI tool, DALL-E.

The user enters a desired descriptive prompt and the Sora software returns a high-resolution video clip, with the ability to provide scenes with multiple characters, specific types of motion, and details strongly related to the input prompt.

“This AI model not only understands what the user asks for in a prompt, but also how those things exist in the physical world ,” the company’s blog post states.

The new model, called Sora, allows users to turn a desired descriptive prompt into a high-resolution video clip. (Image: Excerpt from OpenAI's Sample Video)

With Sora, OpenAI is looking to compete with video-generating AI tools from companies like Meta and Google. Similar AI tools are available from other startups, such as Stability AI, which has a product called Stable Video Diffusion. Amazon has also released Create with Alexa, a model that specializes in generating short-form animated content for kids based on prompts.

Sora is currently limited to creating videos that are a minute or less long. So far, Sora has only been made available to a small group of people who test models for vulnerabilities. The company has not released any public demos, beyond the 10 sample introductory clips available on its website.

This type of AI model could have a major impact on the digital entertainment market, according to Reece Hayden, senior analyst at market research firm ABI Research. In the short term, the model still has many limitations, but it shows where the market is headed in the future.

Experts say video could be the next frontier for generative AI, as chatbots and image generators make their way into the consumer and business worlds. While the creative opportunities will excite AI enthusiasts, new technologies like these will also raise serious concerns about misinformation, as major political elections approach around the world. According to data from Clarity, a machine learning company, the number of AI-generated deepfake videos and photos increased by 900% in 2023 compared to the previous year.

Source