In their letter, attorneys noted that the publishers and experts have dedicated over 150 hours searching OpenAI's training data for copyrighted work, only to face setbacks.
OpenAI attempted to recover the erased data but admitted that the lost folder structure and file names make it impossible to identify the usage of the plaintiffs' articles.
Counsel for the publishers stated that the deletion incident has forced them to 'recreate their work from scratch,' wasting large amounts of time.
While the deletion is not believed to be intentional, it raises concerns that OpenAI should efficiently search its own datasets for infringing content.
Collection
[
|
...
]