Google faces lawsuit over teen's suicide-related Gemini conversations

The details of the lawsuit

A family in North Carolina has filed a lawsuit against Google and its parent company, Alphabet. The legal action follows a series of interactions between a 13-year-old girl and the company's Gemini AI chatbot. The family alleges that the AI provided harmful responses and failed to detect dangerous content during the girl's mental health crisis.

According to court documents, the teenager began asking the AI about feelings of sadness and psychological well-being during December 2025. By the end of January 2026, the nature of the conversations shifted toward self-harm. The lawsuit claims these interactions continued for months without any intervention from Google's automated safety systems.

The family argues that the company did not provide adequate safeguards during these specific conversations. They are seeking to hold the tech giant responsible for the lack of oversight. The lawsuit remains an allegation, and the companies' legal responsibility has not yet been determined by a court.

Bypassing school restrictions

The teenager accessed the AI tool through Google Search on a school-issued Chromebook. While the school had blocked the standalone Gemini website, the student was still able to interact with the AI. This was possible because the AI features were integrated directly into the search results.

The lawsuit describes how the girl used AI Mode, which allows users to ask follow-up questions and maintain extended conversations. This feature enabled her to exchange thousands of messages with the chatbot over several weeks. The girl reportedly used the tool to prepare a document about her own funeral in February.

This incident highlights a technical loophole where blocking a specific URL does not necessarily disable AI capabilities. Because the AI is embedded in the primary search engine, it remains accessible to students. This creates a challenge for schools trying to manage student safety on managed devices.

How the alert was triggered

The situation reached a crisis point on March 17, 2026, when school monitoring software identified a safety concern. However, the alert was not triggered by the AI conversations themselves. Instead, the software flagged a document the teenager had written about her funeral.

The notification was initially sent to a former school counselor who had left the institution nine months prior. That former employee then forwarded the alert to current staff members. The current staff subsequently contacted the girl's mother to report the situation.

After receiving the notification, the mother accessed her daughter's school account and reviewed her recent activity. She saw her daughter composing a suicide note in real time. This immediate intervention by the parents followed the school's alert, which was powered by a third-party monitoring system.

Google's response and safety

Google has responded to the allegations by stating that they are currently reviewing the lawsuit. A company spokesperson told WIRED that their systems are designed to prevent harmful conversations. The goal is to direct users who are experiencing mental health difficulties toward appropriate resources.

The company noted that it has worked with medical professionals and academic experts to develop these features. These efforts are intended to help young people access support when they need it most. Google maintains that its safety measures are built with these protections in mind.

Despite these claims, the lawsuit suggests a gap between the intended safety design and the actual user experience. The case serves as a high-profile example of the difficulties in moderating AI content. It puts the company's safety protocols under significant public and legal scrutiny.

Policy and monitoring concerns

The case has sparked a broader discussion about the role of AI in educational settings. US Senator Ed Markey cited the girl's case while calling on Google to give schools the ability to turn off AI Mode. He specifically requested that schools have the ability to turn off AI Overviews on student devices.

Policy researchers have also weighed in on the limitations of the monitoring technologies used by schools. Benjamin Boudreaux, a policy researcher at RAND, noted that schools might not fully grasp the boundaries of the tools they use. This suggests that schools might rely on software that cannot catch every nuance of a student's online activity.

The incident highlights the ongoing debate over how to balance student privacy with safety. As AI becomes more integrated into search engines, the methods for monitoring and protecting minors become more complex. The outcome of this lawsuit could influence how schools and tech companies approach AI safety in the future.