The group of artists suing Google, including photographer Jingna Zhang and illustrators Sarah Andersen, Hope Larson, and Jessica Fink, discovered that their copyrighted work was used in Google's Imagen image generator's training data, along with billions of other images, through the publicly-available LAION-400M dataset.
The artists argue that AI companies exploiting datasets containing copyrighted material for training purposes are not held accountable due to a gap in the law that fails to address this specific form of copyright infringement, allowing these companies to justify the use of copyrighted work as 'fair use.'
Google faced accusations of purposefully omitting details about its training data source in an update to Imagen to avoid potential legal implications, indicating a conscious effort to sidestep potential copyright infringement claims.
[
add
]
[
|
|
...
]