OpenAI's Sora AI Video Generator Is Mind-Blowing

Imagine a tool that could turn your simple ideas into short, amazing videos. Well, it’s no longer just imagination! OpenAI, the creator of powerful AI tools like ChatGPT and Dall-E, has announced Sora, their first-ever video generator.

The Sora AI video generator is a diffusion model with transformer architecture, meaning its neural network functions in a similar way to ChatGPT. “Text-to-video” means that Sora takes text prompts and turns them into short video clips. The batch released by OpenAI this week is truly breathtaking.

Sora is able to generate complex scenes with multiple characters, specific types of motion, and accurate details of the subject and background.

Let’s see some examples

Want to see Sora in action? Check these out:

These are just a few examples, and they already look way better than what other video AIs can do right now! At first glance, you won’t even assume that these are not real videos.

The risk of misuse

OpenAI and others in the tech field are worried about how video creation tools can be misused. These tools can make very realistic videos that might spread false information or create fake videos of real people. This is a big issue on social media platforms like Facebook and Twitter (now X), which have struggled to stop misinformation and hate speech.

With new tech, these problems could get worse, especially during important election times.

We can’t expect OpenAI to release Sora to the public unless they are sure they won’t cause any public uproar.

Is Sora ready for use?

Right now, OpenAI only showcased what SORA is capable of but there is no way to request access at the moment.

For the latest info on Sora, please refer to: https://openai.com/sora

All the content that’s been going viral on the internet over the past 24 hours has come out of the announcement blog post published by OpenAI.

Rival tools

If this is your first time to hear about AI video generators, it’s important to know that Sora is not the first.

• Runway Gen2: RunwayML is an advanced AI video editing tools, generative video model in research phase (not yet public). https://runwayml.com/
• Stable Video Diffusion: Stability AI announced last year November, but said it was “exclusively for research” and not ready for “real-world or commercial applications at this stage”.
• Pika Labs: Generative AI video tool now open to everyone after a waitlist period. https://pika.art/
• Lumiere: Google’s video model demoed by Google Research, public release timeline unclear.
• Make-A-Video: Meta launched in 2022, known for initially producing blurry, distorted results. Research continues.

Have you checked all the sample videos from OpenAI or Sam Altman? What do you think of Sora?

‍