LinkedIn, the professional networking platform owned by Microsoft, has come under scrutiny for reportedly collecting user data without prior notification. In response to the backlash, the company has revised its terms of service. Nevertheless, it continues to automatically enroll users in data collection practices unless they manually disable the option. This decision has sparked criticism across social media, where many users have voiced their concerns. The updated policy emphasizes that user data will be utilized to enhance artificial intelligence features, such as writing suggestions and personalized post recommendations.
LinkedIn Allegedly Trained AI Using User Data Without User Notification
A report by 404 Media brought attention to LinkedIn’s data collection practices prior to its policy update. Users have reported finding a newly added option within their settings concerning data usage for AI training.
Utilizing user data for training AI models is not an uncommon practice. Companies like Meta have previously acknowledged employing publicly available user content for their in-house Llama models, while Google modified its privacy policy last year to include the training of AI models on publicly accessible web data. The significant issue with LinkedIn, however, lies in the timing of their data collection, which commenced before informing users through official channels.
Typically, platforms are expected to notify users beforehand about such data collection initiatives, allowing them the opportunity to opt-out and safeguard their privacy. LinkedIn’s failure to adhere to this standard has prompted numerous internet users to express their disapproval on various social media platforms, including significant discussions about the implications of the company’s actions.
In light of the situation, LinkedIn has since updated its policy, which now states that “LinkedIn or its affiliates train or fine-tune generative AI models used to create content, including content that may be distributed or made available on LinkedIn’s platform.” The updates specifically mention the training conducted for features like the AI writing assistant and post recommendations.
In a previous communication with TechCrunch, LinkedIn asserted that it employed privacy-enhancing techniques to minimize the collection of personal information during the AI training process. Measures taken reportedly included redacting and omitting personal and identifying data.
To prevent LinkedIn from collecting data through the platform, users can visit this link to access the relevant settings and disable it. However, it remains unclear whether the platform will also eliminate any previously gathered data from the AI models’ training datasets.
