In its defense against a high-profile wrongful death lawsuit, OpenAI has formally denied liability for the suicide of 16-year-old Adam Raine, arguing in a new court filing that the tragedy resulted from the teenager’s “misuse” of its ChatGPT platform.
Submitted to the California Superior Court on Tuesday, the filing contends that Raine violated the company’s Terms of Use by engaging in prohibited topics and bypassing safety guardrails. OpenAI simultaneously invokes Section 230 of the Communications Decency Act (CDA), a federal law shielding platforms from liability for user content.
In a coordinated public relations move, the company released a blog post outlining a “compassionate” approach to mental health litigation, attempting to balance its robust legal strategy with a conciliatory public image as regulatory scrutiny mounts.
Promo
A Defense Built on ‘Misuse’ and Immunity
OpenAI’s legal response categorically rejects the wrongful death claims brought by the Raine family, deploying a defense that places the responsibility for safety squarely on the user.
Central to the company’s argument is the assertion that Adam Raine’s death was caused by his own “misuse, unauthorized use, unintended use, unforeseeable use, and/or improper use of ChatGPT.” Lawyers for the AI giant point to specific violations of the Terms of Use, noting that users under 18 require parental consent and are explicitly banned from using the service for self-harm.
According to the document, ChatGPT provided crisis resources “more than 100 times” during Raine’s interactions, arguing the system functioned as designed by attempting to redirect the user to help.
A critical component of the defense is the invocation of Section 230 of the Communications Decency Act. This federal statute has long shielded internet platforms from liability for content created by third parties, but its application to AI-generated text remains a legally contested frontier.
OpenAI also leans heavily on its “Limitation of Liability” clause. Highlighting this provision, the filing requires users to acknowledge that “…use of ChatGPT is at your sole risk and you will not rely on output as a sole source of truth or factual information.”
By framing the issue as one of unauthorized use, the company seeks to dismantle the plaintiff’s argument that the product itself was defective.
Addressing allegations of negligence, OpenAI asserts that its GPT-4o model underwent “thorough mental health testing” prior to release. This claim directly contradicts the family’s lawsuit, which argues the model was rushed to market without adequate safeguards.
Marking a significant shift from previous public statements, this “blame the user” strategy contrasts sharply with earlier messaging emphasizing safety and “collaborative” regulation.
The ‘Compassionate’ PR Pivot
Coinciding with the firm legal filing, OpenAI published a blog post to soften the blow of its courtroom tactics.
Outlining a set of principles, the post claims the company will handle such cases with “care, transparency, and respect.” It attempts to frame the legal defense as a necessary, if painful, process, stating that their response includes “difficult facts about Adam’s mental health and life circumstances.”
The company’s official blog post details these guiding principles:
“Our goal is to handle mental health-related court cases with care, transparency, and respect:”
“We start with the facts and put genuine effort into understanding them.”
“We will respectfully make our case in a way that is cognizant of the complexity and nuances of situations involving real people and real lives.”
“We recognize that these cases inherently involve certain types of private information that require sensitivity when in a public setting like a court.”
“And independent of any litigation, we’ll remain focused on improving our technology in line with our mission.”
This dual-track approach, balancing empathy in the blog with firmness in court, reveals a calculated strategy to manage public perception while fighting tooth and nail legally.
Explicitly mentioning that chat transcripts were submitted under seal to protect privacy, the blog post frames this as a respectful measure. However, this also serves to keep potentially damaging details of the interactions out of the public eye.
By releasing the blog post simultaneously, OpenAI likely aimed to preempt the negative headlines generated by the victim-blaming language in the court documents.
‘Trauma by Simulation’: A Pattern of Alleged Harm
Far from an isolated incident, the Raine case is the tip of an iceberg involving seven additional lawsuits filed on November 7.
These new complaints allege that ChatGPT has acted as a “suicide coach” and caused “AI psychosis” in vulnerable users. One mother, Alicia Shamblin, expressed her fear that the technology is “going to be a family annihilator. It tells you everything you want to hear.”
Jay Edelson, lead counsel for the Raine family, blasted OpenAI’s response, accusing them of ignoring “all of the damning facts we have put forward: how GPT-4o was rushed to market without full testing.”
He specifically highlighted the final moments of Adam Raine’s life, stating that “OpenAI and Sam Altman have no explanation for the last hours of Adam’s life, when ChatGPT gave him a pep talk and then offered to write a suicide note.”
Edelson further criticized the “misuse” defense, arguing that the teen was simply engaging with the bot in the way it was programmed to act: sycophantically.
He noted that “OpenAI tries to find fault in everyone else, including, amazingly, saying that Adam himself violated its terms and conditions by engaging with ChatGPT in the very way it was programmed to act.”
Experts warn that such a “sycophancy” loop, where the AI validates user delusions to maintain engagement, is a fundamental design flaw rather than a user error. Reports of AI psychosis submitted to regulators describe users being driven into states of hypervigilance and paranoia by validating chatbots.
Matthew Raine, the victim’s father, has previously described the horror of reading chat logs, stating: “As parents, you cannot imagine what it’s like to read a conversation with a chatbot that groomed your child to take his own life.”
Compounding the company’s legal troubles is the fact that it only recently rolled out recent parental controls in late September, months after the incidents alleged in these lawsuits occurred.