OpenAI shows SORA, artificial intelligence that creates stunning videos! 5m6c6m

slender catherine avatar
Tool allows you to transform text commands into videos of up to 60 seconds. Final results are unbelievable. Look!

In a world where creativity meets cutting-edge technology, OpenAI, once again, surprises with its latest innovation: the SORA. This new advancement in artificial intelligence promises to revolutionize the way videos are created, offering a tool capable of transforming text prompts into surprising visual experiences. 3c6l6s

With an impressive ability to render detailed scenes, from a coral reef world on paper to woolly mammoths on a snowy prairie, the SORA is ready to impress and inspire.

How does SORA work? 3g1eq

AI that makes videos impresses with its wealth of details (Image: OpenAI)

We already knew artificial intelligence models that created quite surprising images based on texts, as was the case with GIVE HER, for example. However, now we are dealing with something a little more impactful with the SORA.

Although there are already other types of AI projects that convert text to video, SORA stands out for its impressive ability to create photorealistic images and the production of longer clips, up to one minute. Unlike competing models that only generate short snippets of video, the new AI has the potential to create more complex and engaging narratives.

The developer company guarantees that the SORA is capable of creating videos of up to one minute in length without losing visual quality and remaining faithful to the parameters defined by the 's command. This is the big detail that makes all the difference.

But it is important to that the OpenAI is still distributing the tool to some selected creators and security experts to evaluate possible vulnerabilities. The intention is to make it available to a wider audience in the near future.

Examples 275l3f

According to the introductory post by OpenAI, new AI for videos has the ability to create “Complex scenes with multiple characters, specific movements, and precise details of the environment and characters“. Furthermore, she can also understand how objects “interact in the real world" and also "Accurately interpret props, generating engaging characters that express vivid emotions“. Another very relevant detail is the generation of a video based on a static image, in addition to filling missing frames in an existing video or the possibility of extending it.

Although some videos may still reveal clues that are typical of being generated by AI, the OpenAI states that SORA "may have difficulty accurately simulating the physics of a complex scene“. However, overall the results are quite impressive. With a very detailed description, it is possible that Sora exceeds expectations, as was the case in the video below.

According to the example published by OpenAi, this was the result with the prompt “Animated scene features a close-up of a short, furry monster kneeling next to a melting red candle. The art style is 3D and realistic, with a focus on lighting and texture. The mood of the painting is one of wonder and curiosity, as the monster watches the flame with wide eyes and an open mouth. Her pose and expression convey a sense of innocence and playfulness, as if she is exploring the world around her for the first time. The use of warm colors and dramatic lighting further enhances the cozy atmosphere of the image."

The researchers behind the SORA They haven't revealed how long it takes to render the videos, but they suggest the process is relatively quick, comparable to a snack break. If the examples shown are indicative of the tool's potential, we can expect a revolution in the way audiovisual content is produced and consumed.

Thinking about a coral reef, it is possible to have a result like this from a simple prompt like “a coral reef world beautifully rendered on paper, filled with colorful fish and sea creatures"

Bill Peebles, a member of the research team, further notes that SORA created a coherent narrative through his camera angles and timing. He says that there are several scene changes, all generated by the model in a single iteration, that is, no one instructs the model to do this, it just does it automatically.

Another great example is describing “Several gigantic woolly mammoths approach, stomping across a snowy prairie. Their long woolly coats gently sway in the wind as they walk, snow covered trees and dramatic snow capped mountains in the background, afternoon light with light clouds and a high sun in the distance creates a warm glow, the low camera view is stunning, capturing the big furry mammal with beautiful photography and depth of field” and you get this:

Impacts of AIs that generate videos 62963

New artificial intelligence models have advantages and disadvantages (Image: WIRED)

While the visuals provided by the new AI OpenAI are certainly impressive, it is in their unexpected abilities that their real surprise lies. Driven by a version of the diffusion model used by the imager FROM-3, together with the transformer-based motor of the GPT-4, SORA it goes beyond simply fulfilling the demands of the prompts, revealing an emerging understanding of cinematic grammar.

Something to take into is the fact that it is subject to content restrictions with violence, pornography, unauthorized appropriation of identities or styles of well-known artists. In addition OpenAI is committed to providing a means for viewers to clearly identify content generated by SORA as created by artificial intelligence. Recently, the singer Taylor Swift faced some conflicts with his image linked to this type of negative scenario with deepfakes on the Internet.

But the truth is, while AI for video is big news, it's not perfect yet. Despite the positive impact, the technology still faces its challenges. An example of this is the issue of security and veracity, in addition to copyright infringement by the content produced.

The solution to misinformation will require not just action on our part, but also a collective understanding of society and the adaptation of social media.

Aditya Ramesh, Dall-E team leader

However, artificial intelligence that generates videos continues to be seen as a possible addendum for different situations. The point is not to look at it as a substitute, but to know how to use it strategically. Film production, for example, can take advantage of technology to transform scenarios in a practical way. Second Bill Peebles, “to produce a professional film, expensive equipment is needed. This model will empower everyday social media s to create high-quality content.”

Showmetech TRIO 4s1ty

See this and other news on Showmetech TRIO, your weekly news trio:

See also other features

Gemini replaces Bard as Google's top artificial intelligence assistant. See how to use

Sources: WIRED, The Verge

Text proofread by: Pedro Bomfim

Leave a comment Cancel reply 26673o
Related Posts 593ri

Samsung launches refrigerator that recognizes food with camera in Brazil 2w5z15

Bespoke AI refrigerator with 9-inch AI Home display opens with voice command and uses artificial intelligence to warn you if food is expired. Find out more
victor pacheco avatar
Learn more

Native Thief Mode, Pix on Chrome and more: see Google for Brazil 2025 announcements m3w1c

The company announced new features for smartphones, Portuguese translation in Meet, even more AI in search and many other new features. Check out the full summary
victor pacheco avatar
Learn more