TikTok has decided to jump on board with the emergence of new text-to-image AI tools. The platform now offers its own, albeit pretty basic, text-to-image AI that you can find inside the TikTok app itself.
The insanely popular video app recently added a new AI feature (thanks, Verge) that lets users create their own images for backgrounds with just a few words.
The company is calling the feature “AI greenscreen” and it’s currently available to all users on the platform.
Here’s how it works: users type in a text prompt with a few words describing the kind of image that they’d like. Then, TikTok’s AI greenscreen will spit out its own interpretation of those words in an abstract image.
After an image is created, users can apply that as the background for their TikTok videos. It could potentially be a huge new feature for creators on the platform.
The feature works similarly to apps like DALL-E and Google Imagen, which have recently exploded in popularity.
That said, TikTok’s AI greenscreen is far less advanced, with mostly abstract and swirly images. These other options provide more photo-realistic interpretations.
The Verge also pointed out a potential benefit of the abstract images that AI greenscreen produces. After typing in some violent, inappropriate prompts, the images produced showed little in the way of detail from the actual prompt.
That could serve as a form of automatic moderation, where users aren’t able to create potentially abrasive pictures due to the abstract style.
The style also hides clear nudity but provides the right thematic design, with fleshy tones and body shape outlines.
TikTok’s new AI greenscreen brings text-to-image software to a much wider audience than before. It will interesting to see what kind of content users create with AI greenscreen. And who will be the next mainstream app to add a text-to-image AI to their platform?
Have any thoughts on this? Carry the discussion over to our Twitter or Facebook.