LinkedIn Updates its Terms of Service to Detail AI Training With Member Data

LinkedIn announced an update to its Terms of Service today to make it clearer that it uses data from its members for AI training. LinkedIn does this by default, but the professional social network does offer an opt-out setting for users who don’t want to see their data scrapped for AI training. The company also isn’t doing any AI training with user data in the EU and Switzerland.

Earlier today, an article on 404 Media pointed out that LinkedIn has been using data from its users to train its AI models by default without making it explicit in its terms of service. The company told the publication that it would update its terms of service “shortly,” and it actually did so.

In a support page about its generative AI features, LinkedIn explains how it uses the personal data of its members to train its AI models. As of today, LinkedIn offers a different set of generative AI features, including a writing assistant that can help users write posts or update their profiles.

“Where LinkedIn trains generative AI models, we seek to minimize personal data in the data sets used to train the models, including by using privacy enhancing technologies to redact or remove personal data from the training dataset,” the company explained. “We also do not currently train content-generating AI models on data from members located in the EU, EEA, or Switzerland.”

The use of personal data for AI training has become a sensitive topic, especially after a company like Meta admitted that it’s been scraping data from adult users in Australia since 2007 without giving them a way to opt out. LinkedIn is doing the right thing by offering an opt-out setting to its users, and you can find it on this page.