Are platforms using what you post to train AI, and what do their terms say?
In July 2025 a file-transfer service rewrote its terms after users read them as a licence to train AI on their files.
Guess first
?
Which one is it?
Some are, and some are the subject of disputes. AI tools called are trained on data from the Internet, often including copyrighted material, and a few platforms have licensed or used what their users post. This lesson describes dated cases and the positions of each side; it takes no side.
One route is a paid deal. In February 2024 Reddit announced a partnership with Google, in a deal worth about $60 million per year, to license its real-time user content to train Google's AI model. Reddit's licensing deals with Google and OpenAI account for approximately 10% of its revenue and expire in 2027.
Another route is a change to the terms. In July 2025 WeTransfer was criticised after changing its terms of service, which some interpreted as giving it the right to use files for AI training. The passage mentioned improving models for . In response WeTransfer clarified its terms by removing any mention of machine learning.
A third is the default setting. In September 2024 LinkedIn suspended its use of UK user data for AI model training after concerns raised by the Information Commissioner's Office. A LinkedIn spokesperson said the company has always allowed users to control how their data is used and had now given UK users an .
Why does this keep happening? A legitimate terms of service agreement is legally binding and may be subject to change, so rewording can matter.
Quiz me
0/3
Recap
What a platform may do with your content depends on its terms, which are binding and can change.
💡 A trick to remember it · Your post is a seed: the terms of service decide whether the platform may plant it in an AI's training soil.
Surprising fact · WeTransfer removed any mention of machine learning from its terms in July 2025 after criticism.
Connects to
- 🏥 What happened when ransomware hit London's hospital blood-test provider in 2024?
- 🇪🇺 What rights does GDPR give you over your personal data?
- 📜 Why does nobody read terms of service, and what are you agreeing to?
- ⚖️ Can AI companies train on copyrighted work? What the lawsuits by newspapers, photographers and authors are about
Sources (7)
No source, no claim. Every fact in this lesson (12 claims) cites at least one of these.