Google unveils realistic videos created by Lumiere AI that could be 'exploited'

GOOGLE has revealed a glimpse of its Lumiere platform in a new video.

Lumiere is a new artificial intelligence-powered video generation model developed by Google.

In a new videoThe service is seen creating great 5-second videos based only on text prompts.

This technology marks a significant leap forward in the field of artificial intelligence.

Lumiere uses a unique approach called Space-Time-U-Net (STUNet) to create realistic and coherent videos in one go.

The technology figures out where things are in a video and how they move and change at the same time.

This translates into smoother, more coherent, and ultimately more credible end products, unlike older methods.

Lumiere also generates 80 frames compared to 25 frames from Stable Video Diffusion, The edge noted.

“We introduce Lumiere – a text-to-video diffusion model designed to synthesize videos that depict realistic, diverse, and coherent motion – a critical challenge in video synthesis,” Google researchers said in an article.

“We introduce a Space-Time U-Net architecture that generates the entire temporal duration of the video at once, through a single pass in the model,” they wrote.

The system can perform a variety of video editing tasks, such as converting still images to video.

It can also create videos in a specific style using reference images or text prompts.

Perhaps most notable is that the platform can create animations by altering specific parts of an image Economic times.

The researchers were not aware that there is any risk of abuse when it comes to the platform.

“There is a risk of abuse in creating false or harmful content with our technology,” they said.

“And we believe it is crucial to develop and apply tools for detecting biases and malicious use cases to ensure safe and fair use.”

Lumiere isn't the only AI platform that poses a risk of abuse; ChatGPT and ElevenLabs have already been abused by cybercriminals.

ChatGPT, an advanced chatbot, has been used by amateur hackers to generate malware.

Meanwhile, ElevenLabs has been used to generate deepfakes of celebrities to spread disinformation.

Deepfakes are synthetic media where the face or voice of a person used in an existing video is replaced with that of someone else.

Most recently, the technology was used to create a fake voice recording of President Joe Biden.