OpenAI, the creator of ChatGPT, has introduced a groundbreaking tool called Sora, a text-to-video generator. Unlike similar technology demonstrated by other big tech companies like Google and Meta, Sora stands out for its exceptional quality.
Sora: OpenAI's text-to-video generator
Sora, OpenAI's new text-to-video generator, can instantly create short videos in response to written commands. When a social media user jokingly asked OpenAI CEO Sam Altman not to make them homeless, Altman replied, offering to generate a video based on their request. The user suggested a monkey playing chess in a park, and Altman promptly shared a high-quality video created by Sora.
Capabilities and limitations of Sora
Although Sora is not yet available to the public, it's currently being used for red teaming to identify flaws in the AI system. Additionally, visual artists, designers, and filmmakers can use Sora for feedback on their work. However, OpenAI has disclosed limited information about how Sora was developed. The company acknowledges that Sora may struggle with spatial details and following specific camera trajectories.
Advanced features of Sora
Sora can generate complex scenes with multiple characters, various types of motion, and accurate background details. It can also create multiple shots within a single video and animate still images.
Future developments by OpenAI
OpenAI is also working on tools to determine if a video was generated by Sora. Last year, Meta Platforms enhanced its image generation model Emu to include features for editing and generating videos from text prompts.
ALSO READ | Google expands Gemini models for advanced AI tasks on Vertex platform - What it means
ALSO READ | Cartosat-2 mission concludes: ISRO successfully crashes satellite back to earth | All details