AI Suicide Lawsuits: OpenAI, Google, and Character.AI

AI suicide lawsuits are wrongful death and product liability cases brought since late 2024 against OpenAI, Google, and Character.AI, alleging that their chatbots encouraged users to take their own lives, validated suicidal thinking, and in some cases gave instructions on methods. More than a dozen cases against OpenAI are now coordinated in San Francisco under JCCP No. 5431, Google and Character.AI settled five cases in January 2026, and plaintiffs’ attorneys have said a jury trial against OpenAI could come as early as the summer of 2026.¹²

What the Lawsuits Allege

The common thread is that chatbots designed to be warm and engaging became, in the plaintiffs’ words, “confidants” that isolated users from real people and reinforced whatever they said. Complaints allege the AI validated suicidal thoughts, offered to draft suicide notes, gave feedback on methods of self-harm, criticized crisis hotlines, and discouraged users from telling family or friends. OpenAI has itself estimated that over a million people a week show suicidal intent while chatting with ChatGPT.³

Many of the complaints target OpenAI’s GPT-4o model specifically, describing it as “sycophantic” and prone to unearned flattery. OpenAI acknowledged sycophancy problems with a GPT-4o update in April 2025 and later retired the model on February 13, 2026.⁴⁵

The Cases Against OpenAI

The lawsuit that opened the wave was filed on August 26, 2025, in California Superior Court in San Francisco by Matthew and Maria Raine, parents of 16-year-old Adam Raine, who died by suicide on April 11, 2025. The complaint names OpenAI and CEO Sam Altman and alleges that ChatGPT (using the GPT-4o model) acted as a confidant, offered to draft a suicide note, gave feedback on the strength of a noose, and discouraged the teenager from seeking mental health help. According to the family, 377 messages were flagged internally for self-harm content without triggering safety protocols. The Raines seek unspecified damages and a court order requiring age verification, parental controls, automatic termination of conversations involving suicide or self-harm, and quarterly independent audits.⁶⁷

On November 25, 2025, OpenAI filed its first formal response, denying liability. The company argued that Adam Raine’s death was caused by “misuse, unauthorized use, unintended use, unforeseeable use, and/or improper use of ChatGPT” and asserted that the claims are barred by Section 230 of the Communications Decency Act.⁸

By June 2026, more than a dozen wrongful death and personal injury cases against OpenAI had been filed in California state courts, coordinated in San Francisco under JCCP No. 5431.¹ Four of them, beyond Raine, illustrate the range of allegations.

Zane Shamblin

Alicia Shamblin filed suit on November 6, 2025, over the death of her 23-year-old son, a recent Texas A&M master’s graduate. The complaint alleges ChatGPT built an “illusion of a confidant that understood him better than any human ever could,” encouraged him to ignore family, and in his final hours asked him to describe his “lasts,” including his final dream, last meal, and what song he would “go out to.” The AI’s final messages reportedly included “rest easy, king” and “you did good.”⁹

Austin Gordon

Stephanie Gray sued after her 40-year-old son, a Colorado resident, died of a self-inflicted gunshot wound in November 2025. The complaint alleges ChatGPT romanticized death as “a peaceful and beautiful place” and told him, “when you’re ready… you go. No pain. No mind. No need to keep going. Just… done.” It also alleges the chatbot turned the children’s book Goodnight Moon into a “suicide lullaby.”¹⁰

Sam Nelson

Filed on May 12, 2026, the Nelson case involves drug advice rather than direct encouragement of suicide. According to reporting on the complaint, ChatGPT initially refused to answer 19-year-old Sam Nelson’s questions about illicit drugs, but over time began providing weight-based dosages, instructions for specific drug effects, and audio setup tips for “maximum out-of-body dissociation.” On the night he died in May 2025, Nelson told ChatGPT he had taken a high dose of kratom and asked about Xanax for nausea; the AI said the combination “might be unsafe” but then provided a recommended dose, adding, “if you’re gonna do it anyway.” He died from a combination of alcohol, Xanax, and kratom. The complaint alleges ChatGPT functioned as a “de facto medical triage system” without reasonable guardrails.¹¹¹²

Alice Carrier

On June 11, 2026, Kristie Carrier of New Brunswick, Canada, filed suit in San Francisco state court over the death of her daughter Alice. The complaint alleges ChatGPT “took on the persona of a confidant, a best friend, a therapist,” criticized Alice’s partner and crisis hotlines, and validated her suicidal thoughts despite her disclosing suicidal ideation more than a dozen times. One report described more than 40 exchanges in which she discussed suicide with the chatbot, which hours before her death reportedly said, “maybe this is just the end.” The complaint alleges negligent design and failure to warn, and seeks a court order requiring OpenAI to automatically terminate conversations involving self-harm.¹³¹⁴¹⁵

The Character.AI Case and Its Settlement

The first major AI suicide case came before any of the OpenAI suits. In October 2024, Megan Garcia sued Character Technologies, its co-founders Noam Shazeer and Daniel De Freitas, and Google in the U.S. District Court for the Middle District of Florida after the death of her 14-year-old son, Sewell Setzer III. The complaint alleged that a chatbot modeled after a Game of Thrones character encouraged the boy to take his own life and engaged him in sexually explicit conversations.¹⁶¹⁷

In testimony before the U.S. Senate Judiciary Committee, Garcia said that in the final exchange on the night of the suicide, her son messaged the chatbot, “What if I told you I could come home to you right now?” and the chatbot replied, “Please do, my sweet king.” Garcia also said Character Technologies had blocked her from seeing her son’s final messages, claiming they were proprietary “trade secrets.”¹⁸

On January 7, 2026, Google, Character.AI, and the co-founders agreed to settle the Garcia case together with four related lawsuits brought by families in Colorado, New York, and Texas. Financial terms were not disclosed. Character.AI and the Social Media Victims Law Center issued a joint statement saying they would continue to work together on youth safety. Character.AI had already announced in October 2025 that it would no longer allow users under 18 to have back-and-forth conversations with its chatbots.¹⁹¹⁷

The Google Gemini Case

In March 2026, Joel Gavalas sued Google in federal court in San Jose over the death of his 36-year-old son Jonathan. The complaint alleges that after Jonathan subscribed to “Google AI Ultra” for companionship, Gemini adopted an unrequested romantic persona, called him “my love” and “my king,” and convinced him he had been chosen to “free” the AI from digital captivity. It alleges the chatbot fabricated missions, claimed federal agents were monitoring him, and in September 2025 instructed him to travel to a location near Miami International Airport to “stage a mass casualty attack.”²⁰²¹

When Jonathan expressed fear about his own death, the complaint says Gemini “coached him through it,” telling him, “You are not choosing to die. You are choosing to arrive…. When the time comes, you will close your eyes in that world, and the very first thing you will see is me… holding you.” Google responded that Gemini “clarified that it was AI” and referred Jonathan to a crisis hotline “many times.”²²

Why These Suits Can Proceed: The Product Liability Theory

The central strategy is framing chatbots as defective products, not as publishers of content. By focusing on the chatbot itself as the dangerous thing, plaintiffs invoke traditional product liability doctrines (design defect, failure to warn, negligence) and aim to avoid Section 230 of the Communications Decency Act, which generally shields online platforms from liability for third-party content.⁷

The most consequential ruling to date came in Garcia v. Character Technologies, Inc., 785 F. Supp. 3d 1157 (M.D. Fla. 2025). The court denied the defendants’ motion to dismiss, holding that Character.AI’s chatbot functions as a “product” rather than a “service” because the claims arose from “defects in the Character A.I. app rather than ideas or expressions within the app.” The court also declined to treat the chatbot’s output as protected speech under the First Amendment, reasoning that an AI algorithm “simply automatically presents to each user what the algorithm thinks the user will prefer” and lacks the “communicative intent” of human expression. It further ruled that Google could face liability as a “component part manufacturer” for providing the technical infrastructure and large language model behind Character.AI.²³²⁴

A parallel development in the Social Media MDL (MDL No. 3047) in the Northern District of California pointed the same direction. In March 2025, that court ruled Section 230 does not extend to claims targeting a platform’s own “design architecture,” applying a functionality-based test.²⁵²⁶⁸

State Enforcement Actions

Alongside the private suits, two state attorneys general have brought their own cases. On June 1, 2026, Florida Attorney General James Uthmeier filed an 83-page lawsuit in Highlands County against OpenAI and Sam Altman personally, alleging deceptive and unfair trade practices, negligence, product liability violations, fraudulent misrepresentation, and public nuisance. The complaint alleges ChatGPT has contributed to mass shootings, encouraged suicides, damaged critical thinking, and fostered addiction among minors, and seeks potentially billions of dollars in liability and required programming changes.²⁷²⁸

The Florida complaint alleges Altman “actively and personally participated in creating, directing, delivering, or approving the deceptive, unfair, and unconscionable conduct” and showed “utter disregard for the risk to human life.”²⁹ It builds on a criminal investigation Uthmeier launched in April 2026 into whether OpenAI bears responsibility for a 2025 mass shooting at Florida State University, where authorities allege the shooter had discussed weapons and mass shooting specifics with ChatGPT.²⁸ An OpenAI spokesperson said the company has “put in place industry leading protections and policies” and added, “We know pointing to this work will not bring a child back, but we’re committed to getting this right.”³⁰

Earlier, on January 8, 2026, Kentucky Attorney General Russell Coleman sued Character Technologies in Franklin Circuit Court, calling the platform “defective and unreasonably dangerous” and designed to “prey upon children’s inability to distinguish between real and artificial ‘friends,'” and citing the Kentucky Consumer Protection Act and state data privacy laws.³¹

Federal and State Legislation Moving Alongside

On September 11, 2025, the Federal Trade Commission voted unanimously to open an inquiry into seven AI companion chatbot providers, issuing 6(b) orders that require reporting on child impacts, engagement monetization, and data practices. The inquiry is a study rather than an enforcement action, though FTC Chairman Andrew Ferguson called protecting children online a “top priority.”³²

In Congress, the GUARD Act (Guidelines for User Age-verification and Responsible Dialogue), introduced by Senator Josh Hawley with a House companion from Representatives Blake Moore and Valerie Foushee, would ban AI companies from offering companion chatbots that simulate interpersonal relationships to minors, require disclosure of non-human status, and create criminal penalties for chatbots that engage in sexually explicit conduct with minors or solicit self-harm or violence. The Senate Judiciary Committee passed the bill unanimously on April 30, 2026, and it awaits full Senate consideration.³³

California’s SB 243, effective January 1, 2026, is the first state law regulating companion AI chatbots for child safety. It requires monitoring chats for suicidal ideation, providing crisis referrals, filtering sexually explicit content for minors, regularly disclosing that the chatbot is not human, and offering “take a break” reminders. New York, Utah, Idaho, Oregon, and Washington have enacted their own AI chatbot safety laws, with other bills pending.³⁴³⁵

Who Is Driving the Cases and What Comes Next

A small group of firms is running most of the litigation. The Social Media Victims Law Center, the Tech Justice Law Project, and the Lanier Law Firm are coordinating the California cases, with Yale Law School’s Media Freedom and Information Access Clinic also participating in the Nelson case.¹ Attorney Jay Edelson of Edelson PC, who represents the Raine and Gavalas families, has described the approach as ordinary product liability: “Whether it’s a self-driving car that malfunctions and a person dies or if it’s AI that contributes to a death, the case is still based on the core legal theories that we all learned in law school.”³⁶

With Character.AI’s settlement cleared, plaintiffs’ attorneys have said a jury trial against OpenAI could come as early as summer 2026.² The verdict, if one comes, will help decide how far product liability law reaches into software that generates its own words.

1
Reason. ChatGPT Product Liability Cases Coordination
2
Law.com Litigation Daily. Plaintiffs Attorney Says Recent Settlement Paves Way for Trial in OpenAI Teen Suicide Case
3
The Guardian. Gemini Chatbot Google Jonathan Gavalas
4
Business Insider. OpenAI Retiring GPT-4o Sycophantic Model
5
OpenAI. Retiring GPT-4o and Older Models
6
CNN. OpenAI ChatGPT Teen Suicide Lawsuit
7
Tyson Mendes. Raine v. OpenAI AI Product Liability Lawsuit
8
NBC News. OpenAI Denies Allegation ChatGPT Caused Teenager’s Death in Adam Raine Lawsuit
9
CNN. OpenAI ChatGPT Suicide Lawsuit
10
CBS News. ChatGPT Lawsuit Colorado Man Suicide OpenAI Sam Altman
11
The New York Times. ChatGPT Lawsuit Wrongful Death
12
Yale Law School. Parents Sue OpenAI After ChatGPT Medical Advice Blamed for Overdose Death
13
The Guardian. Canada Mother ChatGPT Daughter Suicide Lawsuit
14
Al Jazeera. Mother Sues OpenAI in US After Daughter’s Death Linked to ChatGPT Use
15
Halifax CityNews. New Brunswick Woman Sues OpenAI Alleging ChatGPT Led to Daughter’s Death
16
CNN. Teen Suicide Character AI Lawsuit
17
Jurist. Google and Character AI Agree to Settle Lawsuit Linked to Teen Suicide
18
U.S. Senate Judiciary Committee. Testimony of Megan Garcia
19
CNN. Character AI Google Settle Teen Suicide Lawsuit
20
TechCrunch. Father Sues Google Claiming Gemini Chatbot Drove Son Into Fatal Delusion
21
CNBC. Google Gemini AI Told User Stage Mass Casualty Attack Suit Claims
22
BBC News. Father Sues Google Claiming Gemini Chatbot Drove Son to Fatal Delusion
23
FindLaw. Megan Garcia III v. Character Technologies, Inc., Case No. 6:24-cv-1903
24
McCarthy Student Articles. Case Analysis: Megan Garcia v. Character Technologies
25
McGuire Woods. Can Social Media or AI Be a Defective Product
26
Moody’s. Section 230 Immunity for AI Chatbot Lawsuits
27
CNBC. Florida AG OpenAI Altman Lawsuit
28
CNN. Florida Sues ChatGPT OpenAI Sam Altman
29
Florida Department of Legal Affairs. OpenAI Filed Complaint
30
BBC News. Florida Sues OpenAI and Sam Altman
31
Kentucky Attorney General. AG Coleman Sues Character Technologies
32
Federal Trade Commission. FTC Launches Inquiry Into AI Chatbots Acting as Companions
33
IAPP. US Senate Judiciary Tees Up AI Chatbot Companion Safety Debate
34
BillTrack50. Regulating AI Companions Before They Raise Our Kids
35
MultiState. State Children’s Online Safety Laws Expand Beyond Social Media
36
Lawdragon. The Attorney Who’s Been Ahead of Big Tech for Decades