The Facts
Here’s the simple truth, typing sensitive financial information into a chatbot is not the same as handing it to a bank teller, a lawyer, or even your closest friend. It feels private, sure. But as one user discovered this week, chatbots do not carry a confidentiality seal, and the digital world is remarkably unforgiving.
On March 31, 2026, a proposed class-action lawsuit was filed in the U.S. District Court for the Northern District of California (San Francisco) against Perplexity AI Inc. and its CEO, Aravind Srinivas. The lead plaintiff, identified in court documents as John Doe, alleges that Perplexity’s chatbot secretly shared sensitive personal and financial information with tech giants Google LLC and Meta Platforms Inc., without user consent.
According to the complaint, Doe entered detailed information during multiple sessions, ranging from family finances to tax and investment details, expecting a private exchange. Instead, he claims, that trust was misplaced. The lawsuit specifically points out that the platform’s so-called “Incognito mode” failed to prevent this tracking, meaning users’ private chats were allegedly exposed even when they thought they were protected.
Perplexity AI has denied wrongdoing. A company spokesperson said that the system does not sell user data and is designed with privacy in mind. No explanation has been provided for how the alleged data transmission occurred, and the chatbot itself, predictably, was unavailable for comment.
The Blame
Let’s be clear, the absurdity here does not come from the chatbot itself. Chatbots are tools, and this one responded exactly as designed, predicting text based on user input. It did not consciously decide to broadcast someone’s personal information. It did not have motives or malice. What it did have was a conversational interface that invites users to type freely, ask follow-up questions, and trust the responses like advice from a human being. And John Doe did exactly that.
The complaint emphasizes that Doe assumed the system functioned like a private confidant. The interface gave no obvious warning that sensitive information could flow through hidden, pre-programmed pipelines or integrations that the user could not see. And when that information allegedly surfaced elsewhere, Doe’s reaction was to point at the chatbot and declare: “This is on you.”
The Real Story
Here’s what actually happened. A human decided to trust a chatbot with highly sensitive data. A human assumed that an interface designed to answer questions in real time was equivalent to a secure vault. And when the outcome wasn’t what they expected, a human cast blame on the tool, rather than reflecting on the choice that led there.
This case highlights a broader trend. The fastest-growing reflex of the 2020s is not to question our own judgment but to blame AI. Something went wrong? AI must have done it. A decision failed? Must have been the algorithm. Human error, laziness, or misjudgment is invisible until it is reframed as the fault of a machine.
In other words, the chatbot became the villain. The people who designed the data pipelines remained invisible. And the human misstep, the fundamental misunderstanding of how digital systems work, was left on the cutting room floor.
The Aftermath
The lawsuit is ongoing. No court has ruled, no settlement has been reached, and there is no evidence of material harm beyond the alleged disclosure. What is clear, however, is that the narrative has already formed. An individual trusted a piece of software, felt betrayed, and decided the AI deserved the blame.
The situation is messy, human, and, yes, a little funny, in the way that seeing a grown adult sue a chatbot can be. Doe’s story may never go viral like a tearful influencer TikTok. However, it still reads as a cautionary tale about misplaced trust, human error, and the absurd lengths we’ll go to to avoid admitting a simple misjudgment.
The Verdict
WHO’S BLAMING AI: John Doe, a Utah man, who entered sensitive personal and financial information into Perplexity AI’s chatbot.
WHAT ACTUALLY HAPPENED: The user typed private details into a conversational interface. According to the complaint, the information allegedly traveled through hidden trackers to third-party platforms Google and Meta, even while “Incognito mode” was active. There is no evidence yet that the chatbot acted with malicious intent.
WHO GOT AWAY WITH IT: Perplexity AI’s systems appear to have functioned as designed. The human choices behind the data policies remain invisible. The chatbot, predictably, is unavailable for comment.
BLAME RATING: 🤖🤖🤖 (3/5 robots) – Classic AI scapegoating: human error came first; the chatbot only delivered what humans instructed it to do.
