—
# Sora: ChatGPT Can Now Generate Photorealistic Videos
In the ever-evolving world of artificial intelligence (AI), it seems like each day brings a significant breakthrough, a new product, or an expansion of existing capabilities. Hot on the heels of Google’s unveiling of Gemini 1.5, the AI experts at OpenAI responsible for the ChatGPT chatbot have announced their latest feat: Sora, the first AI model capable of creating realistic video sequences from simple text inputs.
As of now, Sora isn’t available for general use, but the example videos published on the project’s webpage are already generating widespread amazement. These video sequences, showcasing scenarios like a woman taking an evening stroll through Tokyo, a jeep navigating off-road trails, or a leisurely walkthrough of an art gallery, were all generated from nothing more than brief text descriptions.
Consider the input for the Tokyo evening walk, which consists of just a few sentences detailing the neon-lit urban backdrop, briefly touching on the woman’s attire of black shoes, leather jacket, and red dress, and then describing her confident stride, the reflective wet street, and pedestrians in the background.
Here’s the prompt in full:
„A stylish woman walks down a Tokyo street filled with warm glowing neon and animated city signage. She wears a black leather jacket, a long red dress, and black boots, and carries a black purse. She wears sunglasses and red lipstick. She walks confidently and casually. The street is damp and reflective, creating a mirror effect of the colorful lights. Many pedestrians walk about.“
Initially, Sora will be made accessible to a select group of artists, designers, and filmmakers to experiment with this new tool. At the same time, OpenAI is assessing the model for risks to prevent misuse.
For anyone fascinated by the potential of AI and the capability of video generation, a visit to the project’s demo site is highly recommended. There, you’ll find a background paper that delves into the development and insights of this video generator, showcasing examples that are almost indistinguishable from real footage.
This landmark achievement demonstrates not just the progress in AI technology but also opens up new avenues for creativity and storytelling, pushing the boundaries of what’s possible with AI-generated content. As Sora and similar technologies continue to develop, we may soon find ourselves in a world where the line between reality and AI-generated content becomes increasingly blurred.
—
source: https://www.ifun.de/sora-chatgpt-kann-jetzt-auch-videos-und-zwar-photorealistisch-227712/
