Meet OpenAI Sora: A Revolutionary AI that Creates Hyper-Realistic Videos from Text Up to 1 Minute Long

Meet OpenAI Sora: A Revolutionary AI that Creates Hyper-Realistic Videos from Text Up to 1 Minute Long

Cellular Stockpile Tech Articles Leave a Comment

OpenAI Sora, su nueva IA capaz de generar un vídeo realista a partir de texto con hasta 1 minuto de duración que es una revolución

The use of AI is becoming increasingly widespread across the population and even in the job market. We cannot deny that artificial intelligence has become a double-edged sword, demonstrating how it is possible to generate all types of content at an incredible speed, while also making us fear for our future in the workforce. After witnessing significant advancements in generating images and text, it’s now time for high-quality AI-generated videos, as OpenAI Sora is here and has amazed us with its initial results.

The era of generative AI began relatively recently, with Midjourney and Stable Diffusion being able to generate images from text prompts. Although the first version of Midjourney produced meme-worthy results and didn’t spark much hope, over time they have proven that better results are possible. Each new version they release can create more realistic and accurate content, while also reducing common errors.

OpenAI dominates the AI sector once again with Sora, their new model capable of creating videos up to 1 minute long

With Midjourney, we had a generative AI that could create realistic images and art, and with ChatGPT, we had the ability to generate text. Every question we ask this chatbot, along with its response, can be considered a form of content creation, as it has been trained on millions of parameters and data and can provide complex answers. After seeing AI’s success in images and text, other media needed to be explored. Recently, we saw Google announce Lumiere, their AI model for creating videos of up to 80 frames with better quality than Stable Video Diffusion (SVD).

Until now, we had only seen the possibility of generating a few seconds of video, but OpenAI is ready to completely change the situation with their Sora AI model, which can create videos up to 1 minute long. Additionally, these videos offer impressive quality, something we never imagined could be achieved with AI alone.

Sora can generate realistic or animated videos with high quality

OpenAI Sora works by generating video from text prompts, so those familiar with AI will have no difficulty creating new content. On the official OpenAI website, we can view over a dozen examples of its capabilities, alongside the prompts used for the generated videos. Here we see that not only can Sora create realistic videos, but they also have very few artifacts or visual errors. It can also be used to create videos that resemble 3D animated movies or simulate a period from the past, such as the California Gold Rush era.

Among the examples, we see several realistically generated drone-camera-like videos, footage of various animals, and hundreds of people in a fictitious Chinese Lunar Year celebration. Truth be told, aside from a few videos that need more work (like the one with the archaeologists and the chair), the majority could appear on YouTube and deceive most viewers into believing they are real.

OpenAI has already stated that Sora is still in development and not yet ready for public release. They first want to implement several vital security measures to prevent misinformation and certain types of content. OpenAI is also developing tools to help detect deceptive content and plans to implement C2PA metadata.

This article first appeared on El Chapuzas Informático: OpenAI Sora, the new AI capable of generating a realistic video from text up to 1 minute long, represents a revolution.


Need Help?

Can't Find What Your Looking For? Just fill out our form and we'll ask our trusted distributors

Contact Cellular Stockpile.

Wholesale Inquiries Only.

Email Cellular Stockpile
Tel: | Whatsapp:

Leave a Reply

Your email address will not be published. Required fields are marked *