Life Buzz News

LinkedIn Has Been Secretly Scraping User Data to Train Its AI Model

By Krishi Chowdhary

LinkedIn Has Been Secretly Scraping User Data to Train Its AI Model

What's worse, all users are opted in for data collection by default. If you don't want your data to be used, you need to opt out manually.

LinkedIn has been using user-generated content from the platform to train its AI models, without informing or seeking consent from the users.

The changes were only made this Wednesday (18th September 2024). Blake Lawit, Senior Vice President and general counsel, talked about the company's practices and said that the privacy policy has been updated.

That's not the only issue. All user accounts are automatically opted-in for AI training. If you don't want them to use your data, you need to opt out manually.

Ideally, people should get an option whether they want to share their data or not. But by turning the feature on by default, a lot of users will unintentionally end up sharing their data.

The platform will collect data about you and your activity, including:

The only silver lining here is that the company is working on minimizing the use of personal, identifiable data. It's even considering using privacy-enhancing technologies to remove such data from the training dataset.

Also, users from the EU, Switzerland, Iceland, Norway, and Liechtenstein are exempt from this scheme, at least for now - thanks to their strict user privacy laws.

However, some of the suggestions might include details of other users. For example, if you ask for suggestions on the best advice one has gotten from your mentors, the LinkedIn AI might also share the names of those users whose data is being used in processing that answer. You can obviously edit out those names, but it's still something to worry about.

If you need a little help with opting out, follow this simple guide:

LinkedIn isn't the only one scraping user data for training AI. In this era of booming artificial intelligence, data has become priceless. Every AI developer needs tons of data to train their systems. The more data they can get their hands on, the more advanced their model will be.

Many tech giants, including Meta, Google, and Discord are all doing the same. Some platforms like Reddit have even licensed deals under which AI companies get to legally use the content on its platform for training purposes.

Previous articleNext article

POPULAR CATEGORY

corporate

8042

tech

9144

entertainment

9727

research

4349

misc

10389

wellness

7565

athletics

10233