"As detailed in a new paper, researchers from the open-source AI platform Hugging Face found that the energy demands of text-to-video generators quadruple when the length of a generated video doubles - indicating that the power required for increasingly sophisticated generations doesn't scale linearly. For instance, a six-second AI video clip consumes four times as much energy as a three-second clip."
"While image generators used the equivalent of five seconds of microwave warming to generate a single 1,024 x 1,024 pixel image, video generators proved far more energy-intensive. To spit out a five-second clip, the researchers found that it takes the equivalent of running a microwave for over an hour. If they're consuming far more power as the length increases, the math doesn't look good."
Text-to-video generators exhibit strongly non-linear energy scaling, with video length doubling causing roughly fourfold increases in energy consumption. Short video clips already demand far more power than single images, with a five-second clip equating to over an hour of microwave energy use versus seconds for an image. These scaling dynamics imply rapidly rising hardware and environmental costs as generated videos grow longer. Efficiency techniques such as intelligent caching, reusing existing generations, and pruning inefficient training examples can reduce demands, but substantial electricity consumption from current generative AI tools remains a major unresolved challenge.
Read at Futurism
Unable to calculate read time
Collection
[
|
...
]