ChatGPT Conversations Enter Courts as Evidence Without Privilege

ChatGPT conversations can now be used as evidence in court, and they carry none of the legal protections that shield discussions with lawyers, doctors, or therapists. Users have increasingly turned to AI chatbots for help with personal questions, legal problems, health concerns, and emotional support—often assuming those exchanges remain private. Those exchanges are not protected.

How Conversations Become Evidence

The records from these exchanges can reveal a user’s intentions, private thoughts, and confidential disclosures. They may be obtained through a phone search, a subpoena, the discovery process in a lawsuit, or a report from the AI company itself. Journalists found that chatbot conversations were mentioned in 12 public court cases within the last two years. The actual number is almost certainly higher because much of the evidence gathered during investigations never becomes public.

One case involved a teenager identified as R.K.C., who in 2023 sued Meta, Snap, TikTok, and YouTube, claiming their platforms caused social media addiction and mental health issues. His defense lawyers obtained his ChatGPT usage history. During an October 2024 exchange, R.K.C. asked the chatbot to clarify a comment his father had made about a possible settlement. He reached a settlement with Snap, TikTok, and YouTube by the end of July and dropped his claim against Meta, citing concerns about the length of the trial. The companies denied his allegations, and Mike Morgan, a partner at Morgan & Morgan, said the ChatGPT messages did not affect the outcome.

Chatbot records are also appearing in criminal cases. In Palm Beach County, Florida, OpenAI contacted the FBI after a ChatGPT user repeatedly outlined plans to harm an ex-girlfriend, according to a police affidavit. The FBI passed the information to local police, who identified the woman and found that the user, Darren Zhou, had sent her anonymous messages after the relationship ended. The conversations corroborated the credibility of the threats, police said. Zhou was arrested in May, charged with stalking and making electronic threats, pleaded guilty this month, and received eight years of probation.

Phone Searches and Civil Lawsuits

In many cases, police don’t need to request records from an AI company because they can simply look at the conversations on the person’s phone. Michael Price, litigation director at the Fourth Amendment Center of the National Association of Criminal Defense Lawyers, said people often consent to phone searches even when they have the right to refuse without a warrant.

Why Courts Will Not Protect These Conversations

Chatbot exchanges lack the legal privilege that applies to conversations with lawyers, doctors, or therapists. A federal judge in New York made this clear in the case of Bradley Heppner, a financial executive who had argued prosecutors should not be allowed to examine his conversations with Anthropic’s Claude about possible fraud defenses. The judge ruled the conversations were not protected because Claude is not a lawyer and Heppner’s lawyers had not directed him to consult the chatbot. Heppner was convicted in May of securities fraud, wire fraud, and other charges.

Unlike conversations with professionals bound by privilege, AI chatbot exchanges are treated as discoverable records. Users should assume anything typed into such a system could eventually be seen by opposing parties, investigators, or courts. OpenAI has stated it uses software to identify conversations that may involve dangerous behavior, after which staff can review them. If reviewers determine there is “an imminent and credible risk of harm to others,” the company may contact authorities. OpenAI’s own transparency report shows a marked increase in government and law enforcement requests, with data from more than 80 accounts disclosed during the second half of 2025—more than four times the number from the same period the previous year.

The problem could grow as AI tools take on more responsibilities and gain access to greater amounts of personal data. Andrew Ferguson, a law professor at George Washington University who studies digital surveillance, said the records kept by chatbots might be just the beginning. “The whole of your life will then be accessible to the police,” Ferguson said. Courts have not yet decided whether they will eventually grant any kind of privilege to conversations with AI.