Perplexity's "Incognito Mode" is a "sham," lawsuit says

2026-04-02 · Source: AI - Ars Technica · Field: Legal & Regulatory — Litigation & Dispute Resolution, Compliance & Risk Management, Legal Technology (LegalTech) · Depth: Novice, medium

Summary

A proposed class-action lawsuit filed by an anonymous user, John Doe, alleges that Perplexity's AI search engine shares "enormous volumes of sensitive information" from user chat sessions with Google and Meta without consent. The lawsuit claims that initial prompts and follow-up questions are always shared, with non-subscribed users' conversations accessible via a URL to third parties. Even Perplexity's "Incognito Mode" is described as a "sham," failing to protect user privacy and allegedly sharing personally identifiable information (PII), including email addresses, with Google and Meta. The complaint details instances of financial and health data being shared, accusing Perplexity of violating state and federal laws by not disclosing its use of ad trackers like Facebook Meta Pixel and Google Ads. The class action covers chats shared between December 7, 2022, and February 4, 2026, seeking substantial fines and injunctive relief.

Key takeaway

For CTOs and VPs of Engineering evaluating AI search integration, this lawsuit highlights critical due diligence requirements. You must thoroughly audit third-party data sharing practices and privacy policy adherence of any AI service, especially concerning sensitive user data and PII. Relying solely on a vendor's stated privacy features like "Incognito Mode" without independent verification poses significant legal and reputational risks to your organization.

Key insights

Perplexity is accused of secretly sharing sensitive user chat data, including PII, with Google and Meta, even in "Incognito Mode."

Principles

• User consent is paramount for data sharing.
• Privacy policies must be transparent and accessible.

In practice

• Verify actual data sharing practices of AI services.
• Scrutinize "incognito" or "private" modes for true privacy.

Topics

• Perplexity AI Search Engine
• Data Privacy Lawsuit
• Incognito Mode Deception
• Ad Tracker Data Sharing
• Personally Identifiable Information

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Legal Professional, AI Product Manager, AI Ethicist

Related on AIssential

• Musk’s xAI fired engineer for raising concerns about Grok chatbot, lawsuit claims — An xAI engineer's lawsuit highlights alleged AI safety negligence and privacy violations by Grok, despite Musk's "safer AI" claims.
• Google is building a lifestyle profiling engine, not a "helpful assistant" — Google's "agentic" AI search is framed as a privacy-violating profiling engine using coerced consent to bypass GDPR.
• OpenAI researcher quits over ChatGPT ads, warns of "Facebook" path — Integrating ads into AI chatbots like ChatGPT creates significant privacy risks due to the sensitive nature of user data.
• # The Ladder Pulled Up Behind Them — The core asymmetry is AI labs freely learn from all but restrict others from learning from them.
• Perplexity's "Personal Computer" brings its AI agents to the, uh, Personal Computer — Perplexity's "Personal Computer" brings AI agents to desktops, enabling local file and app manipulation with safeguards.
• A Court Has Ruled That Google Is Liable for False Statements Generated by AI Overviews — A German court ruled Google liable for false statements from its AI Overviews, setting a precedent for AI accountability.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI - Ars Technica.