The newly unredacted court documents show that Meta potentially used pirated materials from Library Genesis to train AI models, raising significant legal and ethical concerns.
An internal Meta employee expressed concern that if the media highlighted the use of datasets from pirated sources, it could damage their negotiating position with regulators.
Despite arguing that their practices fall under 'fair use,' Meta's internal communications reveal that employees were aware of the potential legal and ethical implications of their actions.
A Meta engineer jokingly noted the impropriety of torrenting from a company laptop, indicating an awareness of the questionable nature of their data sourcing practices.
Collection
[
|
...
]