The program presented in networks by the company is able to create high quality clips from simple requests through text

OpenAl, still riding the wave of fascination with ChatGPT and generative Al language models, has unveiled a stunning and revolutionary video creation tool named Sora. This tool can generate video clips just from a description. The announcement was made by OpenAl’s CEO, Sam Altman, on social network X. Sora can produce entire videos at once or extend them to make them longer. It can generate videos just from text instructions, animate a still image, or expand an existing video. The tool is currently in the hands of ‘red teams’ who are tasked with finding potential issues and risks. It is also being tested by visual artists, designers, and filmmakers who are providing feedback on improvements. Despite its impressive capabilities, OpenAl admits Sora still has some evident flaws, such as difficulty simulating complex physics and misunderstanding spatial details. The company is taking precautions before making the tool available to the public, including developing tools to help detect deceptive content and using powerful image classifiers to ensure all generated videos comply with their usage policies. OpenAl is also reaching out to policymakers, educators, and artists worldwide to identify positive use cases for this new tecnhology.