WeTransfer denies using user files for AI training after backlash: Details

WeTransfer updates its terms of service to clarify it doesn't use uploaded files for AI model training, following user concerns

Dutch file sharing service WeTransfer has issued a clarification after a recent update to its terms of service sparked backlash online. The file-sharing platform confirmed that it does not use user-uploaded content to train artificial intelligence (AI) models and has revised the language in its policy to avoid further confusion.

 

In a statement to BBC News, a WeTransfer spokesperson said, “We don’t use machine learning or any form of AI to process content shared via WeTransfer, nor do we sell content or data to any third parties.”

 

The original clause, which drew criticism on social media, had mentioned the potential use of uploaded content to "improve performance of machine learning models that enhance our content moderation process." It also stated WeTransfer could "reproduce, distribute, modify," or "publicly display" user files.

 

WeTransfer said the clause was intended to allow for AI-assisted content moderation and harmful content detection, not to grant rights for AI training or content resale. However, the wording led some users to believe their files could be shared with AI companies or used in training datasets.

In response, the company updated the relevant section on July 15. The revised Clause 6.3 now reads: "You hereby grant us a royalty-free license to use your Content for the purposes of operating, developing, and improving the Service, all in accordance with our Privacy and  Cookie Policy."

 

The updated terms will go into effect for existing users on August 8.

 

The incident mirrors a similar situation from December 2023, when rival service Dropbox had to issue its own clarification after users raised concerns about AI usage of their uploaded files. Like WeTransfer, Dropbox denied using user content for model training.