Meta debuts ‘Make-A-Video’ AI system that creates videos from nothing but TEXT claims

Meta has debuted an AI system that generates videos from nothing but TEXT claims – with some scary and surreal results

  • Meta’s new AI can create videos from nothing but text
  • The AI ​​system, called Make-A-Video, is trained on a mixture of suspense images and unnamed videos
  • So far, it produces impressive results but often looks intimidating or surreal
  • Some of the videos include a disguised spaceship landing on Mars, a ballerina dancing over a skyscraper, and an astronaut floating in space.

Ads

Meta has launched a new type of art-generating AI that can create videos from nothing but text prompts – with frightening, surreal, and impressive results.

The artificial intelligence system announced Thursday, called Make-A-Video, uses existing images with captions to learn about the world and how it is described and uses unlabeled videos to determine how the world moves.

The resulting videos span a gradation from surreal to flowery and scary to compelling.

Mark Zuckerberg hasn’t announced when the system will be available to the public or if there are any restrictions – but there is a registration form that people can fill out if they want to test Make-A-Video in the future.

Meta has launched a new type of art-generating AI that can create videos from nothing but text prompts – with frightening, surreal, and impressive results.

The artificial intelligence system announced Thursday, called Make-A-Video, uses existing images with captions to learn about the world and how it is described and uses unlabeled videos to determine how the world moves.

The artificial intelligence system announced Thursday, called Make-A-Video, uses existing images with captions to learn about the world and how it is described and uses unlabeled videos to determine how the world moves.

Advances in using artificial intelligence to create video-on-demand poses all kinds of ethical dilemmas — not to mention the potential for deep fakes and misinformation.

Meta noted that all content generated by the new system contains a watermark “to help ensure viewers know the video was created using artificial intelligence and not a captured video.”

“Our goal is to eventually make this technology public, but for now we will continue to analyze, test and experiment with Make-A-Video to ensure that every step of the release is safe and intentional,” the tech giant says.

One of the videos shows a fairly realistic looking spaceship landing on the surface of Mars.

Visualize the last astronaut swimming in space from different perspectives.

Mark Zuckerberg hasn't announced when the system will be available to the public or if there are any restrictions - but there is a registration form that people can fill out if they want to test Make-A-Video in the future.

Mark Zuckerberg hasn’t announced when the system will be available to the public or if there are any restrictions – but there is a registration form that people can fill out if they want to test Make-A-Video in the future.

Meta indicated that all content generated by the new system contains a watermark

Meta noted that all content generated by the new system contains a watermark to “help ensure viewers know the video was created using AI and not a captured video.”

Mark Zuckerberg's Meta pumps billions of dollars into creating virtual worlds where people will shop, work and have fun

Mark Zuckerberg’s Meta pumps billions of dollars into creating virtual worlds where people will shop, work and have fun

Two other videos look a bit more surreal.

One shows a ballerina performing a “beautiful and challenging dance on the roof of a very tall skyscraper” with an illuminated view of the city behind.

Another picture shows a blue rhino “flying over a mysterious land”.

“Meta AI is committed to developing responsible artificial intelligence and ensuring the safe use of this state-of-the-art video technology,” the company wrote.

“Our research takes the following steps to reduce the creation of harmful, biased, or misleading content.”

“Across all aspects, spatio-temporal resolution, text fidelity, and quality, Make-A-Video puts state-of-the-art technology into text-to-video creation, as determined by both qualitative and quantitative metrics,” the researchers write in a paper explaining the system.

Previous AI systems that generate static images from text have produced strange results.

Craiyon AI, a different popular text-to-image system not from Meta, has created several different images of what people might look like if all humans joined the metaverse.

Each has an augmented reality headset embedded with its face.

These images were first created when The US Sun AI asked these questions about humans in the metaverse.

A number of technology companies, including Zuckerberg, are pouring tens of billions of dollars into creating virtual worlds where people will be able to shop, work and play.

The images come after a different AI working in the same way produced bizarre images of the ‘last selfie ever taken’ that showed horrific scenes of people in front of nuclear explosions, with rotting flesh and complete destruction behind.

Craiyon AI, a different popular text-to-image system not from Meta, has created several different images of what people might look like if all humans joined the metaverse.  Each has an augmented reality headset built into its face

Craiyon AI, a different popular text-to-image system not from Meta, has created several different images of what people might look like if all humans joined the metaverse. Each has an augmented reality headset built into its face

Ads

Related Posts

Leave a Reply

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