LinkedIn is in hot waters over a lawsuit by its own premium subscribers. The company is facing allegations of sharing personal data of customers with third parties to train large language models (LLMs) without their consent, according to a report by Reuters. In fact, premium customers allege that the Microsoft-owned company shares their private messages as well for the training purpose of AI models.
This lawsuit was initiated in a federal court located in San Jose, California. It specifically represents LinkedIn Premium customers who had sent or received InMail messages. The ‘InMail’ feature allows direct communication between LinkedIn users and is often used for professional networking or outreach. The lawsuit focuses on instances of data sharing that occurred before September 18.
The allegation becomes more significant since, as of January 2025, LinkedIn had over 175 million premium users across the globe. Citing the case filing, the report suggests that this lawsuit accuses the social media company of covertly implementing a privacy setting in August 2024 that allows users to control the sharing of their personal data. After a month, LinkedIn also updated its privacy policy to permit data use for training AI models.
The outrage among premium subscribers increased as LinkedIn provided a “Frequently Asked Questions” link explaining that if users chose to opt out of data sharing, it would not affect any personal data already used for AI training in the past.
The plaintiffs allege that LinkedIn violated the terms of its agreement with users, such as privacy policies or user agreements. They claim that LinkedIn engaged in business practices that are considered unfair or deceptive under California state law. Additionally, the lawsuit is asking for $1,000 per person for each violation of the federal Stored Communications Act, which protects the privacy of electronic communications.
Speaking of LinkedIn’s efforts in the growing AI race, the company added several AI features to the platform, including AI-generated messages in LinkedIn Recruiter. This feature allows users to create personalized InMail messages for potential candidates.
However, Microsoft’s LinkedIn is not the only one to face such allegations. Earlier in November 2024, an Australian Senate inquiry accused tech giants like Amazon, Google, and Meta of exploiting Australian data to train their artificial intelligence (AI) models.
This becomes more interesting as recently Microsoft-backed OpenAI, Oracle, and SoftBank announced a huge investment plan (~ $500 billion) to build AI infrastructure in the United States. In fact, at the start of this year, Microsoft confirmed its massive investment plans to create AI data centers. The company announced that it will invest $80 billion in fiscal 2025 to construct data centers capable of supporting AI workloads.
