Cases › leaked private data › LR-2026-0530

Lawsuit says Perplexity Incognito Mode is a sham

Preliminary Why: 1 report: Ars Technica · subject to change · 1 independent report

Developing: a major or severe case captured in the last 72 hours. Later changes are logged on the corrections page.

Status: no action reported

Severity major

Reported 2 Apr 2026

No source states: loss or records, developer and model, agent tool, could do, deployer, industry.

Severity major: Test met: unbounded scope stated. The complaint alleges the sharing happened to every user, and that chats reached Meta and Google (two third parties). These are allegations in a lawsuit, not established facts.

Perplexity's AI search engine lets users run chat sessions, and a proposed class action filed by an anonymous user alleges that these chats were often shared in full with Google and Meta without users' knowledge or consent. The complaint says opening prompts and clicked follow-up questions were always shared, and that a shared URL could let third parties access the entire conversation. It further alleges that chats were shared along with personally identifiable information even when users turned on Perplexity's Incognito Mode, which the lawsuit calls a sham. The suit targets Perplexity, Google and Meta. These are allegations in a lawsuit; the sources do not describe any court ruling, any response from the companies, or any technical fix.

What happened

1 · What went wrongThe complaint alleges that initial prompts were shared through a URL that third parties such as Meta and Google could use to access the whole conversation.

2 · What it reached or didThe complaint alleges that Incognito Mode did not stop conversations from being shared with Meta and Google, including for paid users.

3 · How it was caughtThe lawsuit says developer tools were used to find that opening prompts and clicked follow-up questions were always shared.

What was done

No source states an action or fix.

Risks

Privacy harm to a person
Primary41 incidents

Read from the sources; not yet reviewed by a person.

Controls

Not yet read for controls. Controls that mitigate the primary risk:

Application

Data minimisation · Sensitive-data checks on outbound data · Human approval for high-impact actions · Output filtering and redaction · People told they are dealing with AI · Risk assessment per deployment

Infrastructure / target

Bot and scraping defences

Evidence

No source states: what the AI was doing, when it happened, when it was found, how many records were exposed, the direct loss, a count of the other parties affected, any court case, the cause, what set it off, what was done.

Show every field, with its source and grade

What happened

Earliest source we hold

2026-04-02

D inferred by us from the described behaviour, reason shown

What went wrong

leaked private data

C one secondary source states it

primary risk

RK15

D inferred by us from the described behaviour, reason shown

Impact

Severity

major

D inferred by us from the described behaviour, reason shown

Who was affected

Every user, subscribed or not, allegedly affected; the complaint alleges enormous volumes of sensitive information from both groups were shared

C one secondary source states it

Kinds of data exposed

email addresses, personally identifiable information

C one secondary source states it

Why this severity

Test met: unbounded scope stated. The complaint alleges the sharing happened to every user, and that chats reached Meta and Google (two third parties). These are allegations in a lawsuit, not established facts.

D inferred by us from the described behaviour, reason shown

Notes on gaps

All harm claims are allegations from a lawsuit, not established facts; the sources do not report a court finding or a company response. The sources name no specific safeguard that held, failed or was missing, so links are empty. No remediation is reported.

LR-2026-0530 · version 1 · published 2026-11-01 after automatic checks; not yet checked field by field by a person · JSON