LinkedIn can have trained AI models using user data without updating its terms.
LinkedIn users within the US – but not within the EU, EEA or Switzerland, probably on account of data protection regulations applicable there – have the choice to opt out. Switch of their settings screen, which states that LinkedIn collects personal data to coach “AI models for content creation.” The switch is just not recent. But first reported by 404 Media: LinkedIn didn’t initially update its privacy policy to reflect data usage.
The terms of service have now been updatedbut often that happens long before a giant change, like using user data for a brand new purpose like this. The idea is that users have the choice to make account changes or leave the platform in the event that they don't just like the changes. This time, it doesn't appear to be the case.
So what models does LinkedIn offer for training? Your own, the corporate says in a question-and-answer session, including models for writing suggestions and post recommendations. But LinkedIn also says generative AI models on its platform might be trained by “one other provider,” comparable to parent company Microsoft.
“As with most features on LinkedIn, whenever you use our platform, we collect and use (or process) data about your use of the platform, including personal data,” the Q&A states. “This may include your use of generative AI (AI models to create content) or other AI features, your posts and articles, your frequency of LinkedIn use, your language preference, and any feedback you may have provided to our teams. We use this data in accordance with our privacy policy to enhance or further develop the LinkedIn services.”
LinkedIn previously told TechCrunch that it “uses privacy protection techniques, including redaction and removal of data, to limit the quantity of non-public information within the datasets used to coach generative AI.”
To opt out of LinkedIn's data scraping, go to the Privacy section of the LinkedIn settings menu on desktop, click on “Data to enhance generative AI,” after which uncheck the “Use my data to coach AI models to create content” option. You also can try opting out more comprehensively via this mannerHowever, LinkedIn points out that any deregistration is not going to affect training courses which have already taken place.
The non-profit Open Rights Group (ORG) has called on the Information Commissioner's Office (ICO), the UK's independent regulator of knowledge protection rights, to analyze LinkedIn and other social networks that train with user data by default. Earlier this week, Meta announced it was resuming its plans to gather user data for AI training after working with the ICO to simplify the opt-out process.
“LinkedIn is the most recent social media company to process our data without asking for our consent,” said Mariano delli Santi, ORG's legal and policy officer, in a press release. “The opt-out model proves once more to be completely inadequate to guard our rights: the general public can’t be expected to observe and track each online company that decides to make use of our data to coach AI. Opt-in consent is just not only required by law, it’s an inexpensive requirement.”
The Irish Data Protection Commission (DPC), the regulator chargeable for monitoring compliance with GDPR, the EU's overarching data protection framework, told TechCrunch that LinkedIn informed them last week that clarifications to its global privacy policy can be published today.
“LinkedIn has advised us that the policy includes an opt-out setting for its members who are not looking for their data for use for training content to generate AI models,” a DPC spokesperson said. “This opt-out option is just not available to EU/EEA members as LinkedIn doesn’t currently use data from EU/EEA members to coach or optimize these models.”
TechCrunch has reached out to LinkedIn for comment. We will update this text if we receive a response.
The demand for more data to coach generative AI models has led more platforms to repurpose or otherwise reuse their vast amounts of user-generated content. Some have even moved to monetize that content—Tumblr owner Automattic, Photobucket, Reddit and Stack Overflow are among the many networks that license data to AI model developers.
Not everyone has made it easy to opt out. When Stack Overflow announced it will begin licensing content, several users deleted their posts in protest – only to see that these posts were restored and their accounts blocked.
