OpenAI has officially launched the AI video-generation tool, Sora, following months of speculation. The release, part of OpenAI’s “12 Days of Ship-mas” campaign, positions the ChatGPT creator as a major player in the generative AI video market, The Verge reported.
Here are five things to know about the new AI video-generation tool.
Sora is OpenAI’s answer to the growing demand for AI-generated video content. Like its predecessor DALL-E, which generates images from text prompts, Sora lets users to type a description of a desired scene and get a high-definition video clip. Sora can also transform still images into video, extend existing clips, and fill in missing frames.
The tool is included in existing ChatGPT accounts, such as Plus and Pro, at no additional cost. It has rolled out to users in the U.S. and most other countries but is still unavailable in Europe and the U.K. because of regulatory concerns.
Sora has generated excitement despite the many challenges. Early users, including revered YouTuber Marques Brownlee, reported issues with realistic physics, with objects sometimes disappearing or passing through one another. Also, Sora restricts prompts involving public figures and copyrighted characters, limiting certain creative applications, Decrypt reported.
OpenAI faces major competition from major tech companies and startups. Tencent recently launched Hunyuan Video, an open-source model it claims outperforms tools like Runway Gen-3 and Luma 1.6. Meta’s Lumiere and Stability AI’s Stable Video Diffusion are also in the space, which is predicted to reach $1 trillion in revenue within the next decade.
Sora’s release raises concerns about misinformation and deepfakes. While OpenAI has emphasized safety testing and red-teaming to address these risks while fostering creative expression, doubts remain, CNBC reported.