Linkedin trains AI with user data

Linkedin, a Microsoft subsidiary, has more than a billion users worldwide Admitted to 404 MediaThe company uses user data to train its AI products. This is especially tricky because the terms of use that should actually be stated don’t currently reflect this.

According to Linkedin, the Terms of Service will change “soon.” That didn’t happen, and from February 2022 members are still offered a user agreement that doesn’t contain any provision allowing the commercial network to use user data to build AI models with the data feed.

According to 404Media, several LinkedIn users discovered the new system on Wednesday, which is said to show that LinkedIn is using user data to improve its creative AI.

Linkedin writers are already feeding AI

Linkedin has already introduced several AI-powered functions for paid premium users, including auto-composing posts or direct messages, or an AI assistant that supports texting. AI also aims to serve as a personal career advisor supporting as a trainer in online courses. “Extremely practical advice from industry-leading entrepreneurs and practitioners” processed and personalized by AI.

Not all AI features are offered globally. AI-powered insights into job ads are currently only available to paying users in the US.

Certain, highly active LinkedIn users are also repeatedly asked to contribute long articles and are attracted by the fact that they are rewarded with badges if they write more often. These articles (Here is an example) is presented by “KI and the Linkedin Community”, said about the authors of these articles, each devoted in detail to a topic area.

See also  Here's how Apple used Google's help to train its AI models

Training with user data is complex

Meta or X (formerly Twitter) is known to use user data to train their AIs. With Meta or X, you can opt out to prevent your own data from being used to train llama or croc models.

For Meta, the matter is so complicated that Mark Zuckerberg’s company has not launched some AI services in the EU – such as his chatbot Meta Ai – for fear of violating the GDPR. It is known that users’ public Instagram and Facebook images were used to create the image of the llama models.

Leave a Reply

Your email address will not be published. Required fields are marked *