LinkedIn Scraped User’s All Info for AI Training Before Updating TOS

LinkedIn might have utilized user data to train its AI models without revising its terms of service.

In the U.S., LinkedIn users have an opt-out option in their settings that reveals the company collects personal data to train “content creation AI models,” but this feature is not available in the EU, EEA, or Switzerland, likely due to stricter data privacy regulations in those regions.

LinkedIn Scraped User's All Info for AI Training Before Updating TOS

While the opt-out toggle itself isn’t new, 404 Media first reported that LinkedIn failed to update its privacy policy to accurately reflect this data usage.

The terms of service have been revised, but typically, such updates happen well in advance of significant changes, like repurposing user data.

This allows users the opportunity to modify their accounts or exit the platform if they disagree with the updates. However, that doesn’t appear to be the case this time.

So, what kind of models is LinkedIn developing? According to the company in a recent Q&A, they are working on their own models, including those for writing suggestions and post recommendations.

Additionally, LinkedIn mentions that generative AI models on its platform may also be trained by “another provider,” such as its parent company, Microsoft.

The Q&A states, “Like many features on LinkedIn, when you interact with our platform, we collect and utilize (or process) data related to your usage, which may include personal information.”

“This could encompass your engagement with generative AI (AI models for content creation) or other AI functionalities, your posts and articles, your frequency of LinkedIn use, your preferred language, and any feedback you’ve shared with our teams. We leverage this data, in line with our privacy policy, to enhance or develop LinkedIn services.”

LinkedIn previously said that it employs “privacy enhancing techniques, such as redaction and removal of information, to minimize the personal data included in datasets used for generative AI training.”

To opt out of LinkedIn’s data collection, go to the “Data Privacy” section in the LinkedIn settings menu on your desktop.

From there, select “Data for Generative AI improvement” and turn off the option labeled “Use my data for training content creation AI models.”

Additionally, you can attempt a more comprehensive opt-out using this form, but LinkedIn clarifies that this will not impact any training that has already occurred.

linkedin-data-toggle

The nonprofit Open Rights Group (ORG) has urged the Information Commissioner’s Office (ICO), which oversees data protection rights in the U.K., to launch an investigation into LinkedIn and other social networks that automatically use user data for training.

Earlier this week, Meta revealed its intention to restart efforts to collect user data for AI training after collaborating with the ICO to streamline the opt-out process.

“LinkedIn has become the most recent social media platform discovered to be processing our data without obtaining consent,” stated Mariano delli Santi, ORG’s legal and policy officer.

“The opt-out model clearly falls short in safeguarding our rights; it is unreasonable to expect the public to keep track of and pursue every online entity that opts to use our data for AI training. Opt-in consent is not only a legal necessity but also a basic requirement rooted in common sense.”

Ireland’s Data Protection Commission (DPC), which oversees adherence to the GDPR—the EU’s comprehensive privacy regulations—communicated to reporters that LinkedIn notified them last week about upcoming clarifications to its global privacy policy, scheduled for release today.

“A spokesperson for the DPC stated, ‘LinkedIn informed us that the updated policy would feature an opt-out option for members who prefer not to have their data utilized for training content-generating AI models.’

However, this opt-out is not accessible to EU/EEA members, as LinkedIn is not presently using data from these regions to train or refine these models.”

The increasing need for data to train generative AI models has prompted an expanding number of platforms to recycle or repurpose their extensive collections of user-generated content.

Some of these platforms have taken it a step further by monetizing this content; for instance, Automattic, the owner of Tumblr, along with Photobucket, Reddit, and Stack Overflow, are now licensing data to developers of AI models.

Not every platform has facilitated an easy opt-out process. When Stack Overflow declared its intention to start licensing content, a number of users responded by deleting their posts in protest, only to find that those posts were reinstated and their accounts suspended.

Stories You May Like

Help Someone By Sharing This Article