OpenAI releases text-to-video mannequin Sora for ChatGPT Plus and Pro customers
OpenAI mentioned on Monday it has launched its synthetic intelligence mannequin, which creates video from textual content, to ChatGPT Plus and Pro customers, increasing its foray into multimodal AI applied sciences.
The Microsoft-backed firm, which kicked off a generative AI craze with the launch of its ChatGPT chatbot in November 2022, goals to focus on comparable text-to-video instruments from Meta and Alphabet’s Google, together with Stability AI’s Stable Video Diffusion.
The AI mannequin, named Sora, was first launched in February, however its entry was restricted to security testers in its analysis preview part. It is now accessible to ChatGPT Plus and Pro customers as Sora Turbo at no further value.
“We’re working on tailored pricing for different types of users, which we plan to make available early next year,” the corporate mentioned in a weblog publish.
Users will have the ability to generate movies as much as 1080p decision, as much as 20 seconds lengthy and in widescreen, vertical or sq. side ratios.
OpenAI mentioned whereas Sora wouldn’t but be accessible in EU nations, Switzerland and the UK, the AI mannequin can be accessible in different areas the place ChatGPT is current.
The firm additionally mentioned it should block the creation and add of damaging types of abuse, reminiscent of youngster sexual abuse supplies and sexual deepfakes, on Sora to stop its misuse.
“Uploads of people will be limited at launch, but we intend to roll the feature out to more users as we refine our deepfake mitigations,” it mentioned.
