Perplexity AI Sued Over Alleged Secret Data Sharing With Meta, Google

Somewhere between the coffee shops and co-working spaces lining Market Street, tens of thousands of San Francisco residents have been typing their most sensitive questions into Perplexity AI's chat interface. A 135-page class-action complaint filed March 31 at the Philip Burton Federal Building on Golden Gate Avenue claims those conversations were streaming in real time to Meta and Google's servers from the moment users logged in.

The lawsuit, Doe v. Perplexity AI Inc., No. 3:26-cv-02803, names Perplexity, Meta Platforms and Google as defendants. The lead plaintiff, a Utah man identified only as John Doe, says he used the AI search engine to research his taxes, investments and family finances, believing those conversations were private. According to the complaint, trackers embedded in Perplexity's code activated at login and transmitted the full text of user conversations to Meta and Google servers before Perplexity itself had even processed the queries. The tracking continued even when users activated their browser's Incognito mode, a feature widely understood to provide at least partial privacy protection.

The complaint accuses the companies of using the intercepted data to "exploit this sensitive data for their own benefit, including targeting individuals with advertising and reselling their sensitive data to additional third parties." The statutory claims span California consumer privacy laws and federal wiretapping statutes, among other theories.

Jesse Dwyer, a Perplexity spokesperson, responded by saying the company had not yet received the filing. "We have not been served any lawsuit that matches this description so we are unable to verify its existence or claims," Dwyer said. Representatives of Google didn't immediately respond to a request for comment. Meta offered no public statement on the filing.

The proposed class covers all free-tier users who chatted with Perplexity and had their information sent to Meta or Google between December 7, 2022, and February 4, 2026, excluding paid "Pro" and "Max" subscribers. A California-only subclass is proposed separately, giving the case particular standing under state privacy law. If the court certifies the classes, the potential plaintiff pool could number in the millions.

The technical claim echoes the structure of what became one of the most expensive privacy enforcement episodes in recent tech history. In a wave of lawsuits from 2022 through 2024, hospitals across the country faced litigation after Meta's Pixel tracker was found transmitting patients' sensitive health-related browsing activity to Meta's advertising infrastructure without their knowledge, producing settlements totaling hundreds of millions of dollars and prompting federal health regulators to issue new guidance on third-party trackers. The mechanism alleged in the Perplexity complaint is structurally identical: a third-party script embedded by the platform operator becomes a passive conduit, funneling user content to advertising companies whose data-collection practices are governed by entirely different privacy standards than the AI tool users believed they were interacting with.

For California residents, the Consumer Privacy Rights Act creates immediate practical recourse. Users who want to understand what Perplexity holds can submit a data access request through the company's privacy settings page. Those who want their information removed can file a deletion request through the same portal. Anyone who has used the platform to research medical, financial or legal matters should export their full chat history before submitting a deletion request, since that export creates a personal record of what was shared. Because the alleged tracking is described as triggering at login rather than during any particular search, even casual users may have had their activity captured.

Perplexity, led by CEO Aravind Srinivas and headquartered in San Francisco, was most recently valued at $22.6 billion in January 2026 after raising $1.72 billion in total funding. The company has aggressively positioned itself as a direct challenger to Google Search, an ambition that now collides with a federal lawsuit described as one of the most detailed privacy complaints against an AI company to date.

The lawsuit joins a wave of regulatory and legal scrutiny facing AI startups over how they collect, store and share user data. Early procedural battles in San Francisco federal court, including motions to dismiss and fights over class certification, will determine whether the case survives long enough to force discovery into Perplexity's actual server-side data practices, the factual core of what the complaint alleges and what millions of users who thought their conversations were private will want answered.