Artificial intelligence is everywhere, and LinkedIn is no exception. The professional social network recently updated its terms of service, announcing that it will use member data to train its generative AI models.
This decision, which now extends to Europe, Switzerland, and the UK, raises questions about privacy and ethics.
The Yiaho team tells you everything you need to know and how to protect your data!
LinkedIn and AI: Data collection… by default!
Starting November 3, 2025, LinkedIn plans to use public data from its members—profiles, posts, comments, job searches—to train its AI algorithms. This initiative, already in place in the United States, aims to improve the user experience.
According to LinkedIn, this data will allow for features like personalized content suggestions, better matching between recruiters and candidates, and assistance with writing messages and posts. However, this collection is enabled by default, meaning your data will be used unless you explicitly take action to object.
LinkedIn specifies that private messages are not affected and that this practice complies with the General Data Protection Regulation (GDPR). However, as noted by the CNIL, the balance between corporate interests and respect for privacy remains a sensitive issue.
Why this collection is raising concerns
Using data to train AI models raises several concerns:
- First, LinkedIn indicates that the opt-out (refusal to participate) is not retroactive. This means that if your data has already been used to train models before you disabled the option, it will remain integrated.
- Furthermore, the lack of transparency regarding the exact use of this data and the actual benefits for users fuels the debate. Does improving the experience justify the massive exploitation of personal data?
This announcement comes within a broader context where other tech giants, such as Facebook and Instagram for Meta AI, have faced criticism for similar practices. Users often find themselves faced with a default choice that prioritizes corporate interests at the expense of control over their own data.
How to opt out of your data being used
LinkedIn is joining the movement of platforms exploiting user data to develop generative AI. While the promises of an improved experience are appealing, they come with ethical and legal questions. Until there is more clarity, it is simple to disable the collection of your data.
LinkedIn offers the possibility to disable this collection, although the option is somewhat buried in the settings. Here are the steps to follow to protect your data:
- Go to your LinkedIn profile and click on Settings Privacy.
- Go to the Data Privacy tab.
- Look for the option Data for AI Improvement.
- Uncheck the box that is enabled by default (except for certain student profiles).
By following these steps, your data will no longer be used to train LinkedIn’s AI models. Good news: disabling this option should not deprive you of future AI features, such as writing assistance!
Our data for training AI: Should we be worried?
Using data to train generative AI has become a common practice in the tech industry. While LinkedIn promises improvements for its users, it is legitimate to question the transparency and limits of this collection. The non-retroactivity of the opt-out is particularly problematic, as it limits users’ control over data that has already been exploited.
For European users, the GDPR provides a protective framework, but its application in this context remains unclear. As the CNIL notes, everything depends on the balance between benefits for the company and respect for user rights. It is up to you to decide if you want to participate in this new era of AI on LinkedIn or protect your data.
Source: LinkedIn
