Perplexity AI Shared Your Private Chats With Meta and Google—Even in Incognito Mode

How the Tracking Works

The complaint describes the trackers as "undetectable" software embedded in Perplexity's code. While many websites use third party analytics and advertising scripts, the concern here is what data those scripts can access. Typical website trackers collect page views and click patterns. The lawsuit alleges these trackers captured the actual content of AI conversations, a far more intimate data set.

The complaint states that Meta and Google used this data to "exploit this sensitive data for their own benefit, including targeting individuals with advertising and reselling their sensitive data to additional third parties."

This is not unusual for web advertising infrastructure. Meta's tracking pixel and Google's analytics scripts are present on millions of websites. What makes this case different is the nature of the data being collected: people use AI chatbots for deeply personal questions they would never type into a search engine.

The Incognito Mode Problem

Perplexity offers an Incognito mode that promises not to save conversations to a user's search history. The lawsuit argues this creates a false sense of privacy. Users who believe their chats are private may share more sensitive information, not realizing that third party trackers are still active and transmitting data to external companies.

This mirrors a pattern seen across the tech industry. Google paid a $5 billion settlement in 2024 over allegations that Chrome's incognito mode still collected user data. The word "incognito" carries an expectation of privacy that the technical implementation often does not deliver.

What the Companies Said

Perplexity spokesperson Jesse Dwyer stated the company "has not been served with any such lawsuit" and could not verify its claims. Meta pointed to its existing policies that "prohibit advertisers from submitting sensitive data." Google did not immediately comment.

Meta's response is notable because it shifts responsibility to the website operator. Meta's advertising policies do prohibit sending sensitive information through its tracking pixel, but enforcement of that policy relies on the website operator filtering the data before transmission. If Perplexity's implementation sent conversation content through Meta's tracking infrastructure, the question becomes whether Meta should have detected and blocked it.

Why AI Chatbots Are a Bigger Privacy Risk Than Search Engines

People interact with AI chatbots differently than traditional search engines. A Google search is typically a few keywords. An AI conversation can be paragraphs long, containing specific details about personal circumstances, medical conditions, legal questions, and financial situations.

When that conversational data gets funneled into advertising networks, it creates an extraordinarily detailed profile. A person asking an AI about tax strategies for a divorce, for example, could find themselves targeted with ads for divorce attorneys and financial advisors, a level of behavioral targeting that feels invasive even by today's standards.

This case highlights a broader concern about the AI industry's business model. Many AI companies offer free or low cost products and monetize through data. If the allegations are true, Perplexity was not just building a search engine; it was building a surveillance pipeline with a chatbot interface.

The Legal Basis

The lawsuit cites violations of California privacy laws, federal computer privacy laws, and fraud statutes. If certified as a class action, it could include every Perplexity user whose data was shared without consent.

California has been increasingly aggressive about enforcing privacy violations related to tracking and data sharing. The state recently fined Ford $375,000 for ignoring opt out requests, and its privacy agency has signaled that companies treating user data as a revenue stream will face consequences.

How to Protect Yourself

Whether or not the lawsuit's allegations are proven, the underlying issue is real: most AI chatbots embed third party tracking. Here is what you can do:

• Use browser extensions like uBlock Origin that block known advertising trackers, including Meta Pixel and Google Analytics scripts
• Never share sensitive personal information (SSNs, financial details, medical conditions) with any AI chatbot
• Do not assume "Incognito" or "private" modes in any application actually prevent third party tracking
• Review the privacy policy of any AI tool you use, paying attention to language about "analytics partners" and "advertising"
• Consider using privacy focused browsers like Firefox or Brave that block third party trackers by default

The Bottom Line

The Perplexity lawsuit represents a new front in the fight for digital privacy. As AI tools become integral to daily life, the data they collect becomes exponentially more valuable and more dangerous. A traditional search history reveals what you are curious about. A chatbot history reveals what you are worried about, what you are planning, and what you do not want anyone else to know.

If you treat your AI conversations as private, you may want to reconsider. The companies behind the curtain are almost certainly listening, and now, according to this lawsuit, they may be sharing what they hear with the biggest advertising companies on earth.