Perplexity AI Is Being Sued for Secretly Sharing Your Conversations With Meta and Google. Here's What's Actually Happening.

On April 1, 2026, a class-action lawsuit was filed in United States federal court in San Francisco accusing Perplexity AI of secretly sharing the personal information of its users with Meta and Google, in violation of California privacy laws. The complaint was filed by a Utah man identified only as John Doe, who is seeking to represent the full class of Perplexity users. According to the Bloomberg report that broke this story, the allegations are specific and disturbing: hidden tracking software embedded in Perplexity's code allegedly transmits the content of user conversations to Meta and Google the moment users log into the platform — and the trackers continue operating even when users switch to Perplexity's Incognito mode.

What the Lawsuit Actually Claims

• Trackers activate immediately on login. The complaint alleges that as soon as users open Perplexity's homepage, trackers download onto their devices and give Meta and Google full access to conversations between users and the AI search engine. This happens before any search is conducted.
• Incognito mode offers no protection. The lawsuit specifically alleges that switching to Perplexity's Incognito mode does not prevent the tracking. Users who believed their searches were private were allegedly being tracked regardless.
• The data is allegedly used for advertising and sold to third parties. The complaint claims Meta and Google use the transmitted conversation data to target users with advertisements and may resell it to additional third parties — a practice the plaintiff calls a 'backdoor' into sensitive AI conversations.
• The plaintiff shared highly sensitive information. The specific harm alleged is not theoretical. John Doe says he shared details about his family's finances, tax obligations, and personal investment strategies with Perplexity — conversations he believed were private.
• Meta and Google are named as defendants. The lawsuit names all three companies — Perplexity, Meta, and Google — for allegedly violating federal and state privacy and fraud laws. All three companies deny the allegations.

What Each Company Says

Perplexity's spokesperson Jesse Dwyer stated: 'We have not been served any lawsuit that matches this description, so we are unable to verify its existence or claims.' This is a technically precise denial — it does not say the tracking does not occur, only that they have not been formally served. Meta referred to its Facebook help page, noting that sharing sensitive information violates company policies. Google has not issued a specific comment. The denials from all three companies do not address the technical mechanism alleged — embedded trackers that operate independently of stated privacy policies.

This Is Not Perplexity's First Legal Problem in 2026

The class-action is the third major legal threat Perplexity has faced in quick succession. Amazon filed a separate lawsuit accusing Perplexity's Comet browser of covertly accessing Amazon customer accounts while disguising automated activity as human browsing. A federal judge has already temporarily blocked Perplexity Comet from accessing Amazon. Reddit accused Perplexity of unauthorized scraping of its content and found that after creating a post only Google could see, the content appeared in Perplexity's results within hours. The Encyclopedia Britannica and Merriam-Webster are suing OpenAI — which Perplexity also faces from the New York Times and Chicago Tribune. The pattern: multiple lawsuits across data scraping, privacy, and account access, all within the same period.

What This Means for Perplexity Users Right Now

• If you have shared sensitive financial, medical, or personal information with Perplexity, treat it as potentially compromised. The lawsuit's claims are unproven but specific. Until the case resolves, assume the worst-case scenario for data you have shared.
• Incognito mode in Perplexity may not protect you. The lawsuit alleges explicitly that tracking continues in Incognito mode. Until Perplexity provides a technical explanation that contradicts this, do not rely on Incognito as a privacy protection.
• Alternative AI search tools exist. For research that needs to be private — financial planning, medical research, legal questions — consider tools where the data handling architecture is more transparent. Perplexity's privacy policy does not clearly describe the tracking alleged in the lawsuit.
• Watch for class membership notifications. If the lawsuit proceeds and a class is certified, Perplexity users may be entitled to participate. Monitor your email for notifications from legal counsel.

Which AI Tools Currently Have the Clearest Privacy Position

Claude (Anthropic): No ads, Constitutional AI training, explicit commitment to not training on user conversations by default. The clearest privacy position of the major AI assistants. Anthropic even ran Super Bowl ads mocking ChatGPT's ad rollout. ChatGPT (OpenAI): Now serving ads to Free and Go tier users. Conversation data used for ad targeting based on topics. Users can opt out of training data use in settings, but the advertising mechanism creates a financial incentive to mine conversation context. Gemini (Google): Integrated into Google's advertising ecosystem. Conversation data subject to Google's privacy policy, which permits use for improving Google's products and services. Perplexity: Now subject to active litigation over alleged undisclosed data sharing. Privacy position currently the least clear of the major players. Local AI (running models on your own hardware): Zero data transmission. Complete privacy. Becoming more accessible with hardware like Intel's $949 32GB GPU.