Google DeepMind | Visualising AI - Multimodality

13 Views

  • Thanks! Share it with your friends!

    You disliked this video. Thanks for the feedback!

cgshort
Published
VISUALISING AI by Google Deepmind x Twistedpoly

Multimodality:
With multimodality the idea was showing different inputs of various shapes and forms being pushed in from above. First we see just a few of them, then a lot gets "thrown" in. These then get a bit organised and lined up and we see them being fed into this abstract object representing AI. What we end up with, is an output as a fantasy world that is also being made of all the inputs we saw in the beginning.

Sound by Marko Lavrin

About Visualising AI
Visualising AI aims to open up conversations around AI. By commissioning a wide range of artists to create open-source imagery, the project seeks to make AI more accessible. It explores the roles and responsibilities of the technology, weighing up concerns and societal benefits in a highly original collection of artist works. 

The ongoing project invites cutting-edge artists to spend time with scientists, engineers, researchers, ethicists and more, discussing key themes in AI, before using those conversations as a catalyst to create new artworks. Crucially, each artist is given complete creative freedom, explicitly briefed to interpret the shared knowledge in whichever way they choose.
Category
Motion Graphics

Up Next