LinkedIn will use your data to train its AI starting November 2025 (unless you opt out)
Starting November 3, 2025, LinkedIn will begin using the profiles, posts, resumes, and public activities of its members to train its AI models.
This change has raised concerns among many users, as it will be enabled by default: you will need to manually opt out if you do not wish for your data to be used.
Opting out, but not retroactively
LinkedIn clarifies that the option can be found in the privacy settings, under “Data for improving generative AI.” The process is simpler than on other social networks, but it only applies to data collected after opting out. Data recorded prior to that will remain in the training environments.
For a complete opt-out, members can also fill out an objection form to LinkedIn.
This measure applies not only to users in the EU but also to those in the EEA, Switzerland, Canada, and Hong Kong. However, LinkedIn confirms that data from users under 18 will be excluded from this process.
A legal framework that favors LinkedIn
Microsoft, the parent company of LinkedIn, justifies the automatic opt-in by citing the notion of “legitimate interest” in law. This allows them to bypass the obligation to obtain explicit consent, as long as an opt-out option is available.
LinkedIn is not alone in pursuing this path. Meta announced a similar initiative for Facebook and Instagram in September 2024. After a pause due to a complaint from the UK regulator (ICO), the company resumed its data collection while making opt-out options clearer.
LinkedIn is following the trend among major social networks: utilizing user data to enhance AI. The platform is attempting to maintain transparency by offering an accessible opt-out; however, the fact that this setting is enabled by default raises genuine concerns regarding trust and respect for privacy.
