Welcome to the forefront of conversational AI as we explore the fascinating world of AI chatbots in our dedicated blog series. Discover the latest advancements, applications, and strategies that propel the evolution of chatbot technology. From enhancing customer interactions to streamlining business processes, these articles delve into the innovative ways artificial intelligence is shaping the landscape of automated conversational agents. Whether you’re a business owner, developer, or simply intrigued by the future of interactive technology, join us on this journey to unravel the transformative power and endless possibilities of AI chatbots.
A growing number of lawsuits aim to hold OpenAI responsible for what plaintiffs say is their chatbot’s role in people’s deaths. Northeastern experts wonder if widespread change will occur because of them.
A growing number of lawsuits are seeking to hold OpenAI accountable in cases where plaintiffs say the company’s ChatGPT chatbot played a role in crimes and deaths.
As the question of liability plays out in court, some experts are unsure whether these legal queries will bring about a massive change in the industry.
“When you get down to it, what is your actual causal theory of how X technology harmed Y individual? It is not that easy to prove that,” said John Wihbey, professor of media and technology at Northeastern and director of the AI-Media Strategies Lab.
“I could be wrong, but I would be surprised if there’s blockbuster case after blockbuster case, as we had in the tobacco settlements, where it sort of fundamentally reshaped an entire industry,” he said, referring to the landmark 1998 Tobacco Master Settlement Agreement that, among other things, restricted tobacco marketing and required indefinite payments to states for smoking-related healthcare expenses.
The recent cases all involve tragic instances of death, and often, suicides. Last December, for instance, the estate of an 83-year-old Connecticut woman who was killed by her 56-year-old-son who then killed himself, sued OpenAI claiming conversations her son had with ChatGPT led to their deaths by murder-suicide.
Two lawsuits from unrelated incidents were filed within days of each other in May. One of these lawsuits came from the family of a victim in the 2025 shooting at Florida State University, claiming ChatGPT guided the accused shooter in carrying out the attack. The other came from the parents of a 19-year-old in Texas who say their son took a fatal mix of drugs upon the advice of ChatGPT.
The allegations that link these cases and others include wrongful death, product design defect and failure to warn. Many point to a specific model of the chatbot called GPT-4o, which the company introduced in May 2024 — two years after the original release of ChatGPT — and retired in February.
OpenAI has denied responsibility, according to news reports about the lawsuit related to the FSU shooting. In May, the company shared information about its safety updates to “better recognize when risk may be emerging over time.” Northeastern Global News reached out to the company for further comment.
‘Everywhere’ and ‘nowhere’
AI companies and their chatbots, which are a type of AI known as large language models, are in a particularly paradoxical position, said Hilary Robinson, associate professor of law and sociology at Northeastern.
A 1996 U.S. law known as Section 230 protects platforms like Google and Facebook from liability for what’s published through third-party content, like search results or social media posts. Then there’s the First Amendment, which, to an extent, also protects free speech.
Robinson said the companies appear to benefit from legal protections on both sides — avoiding publisher liability while also claiming free speech protections.
“At present, they’re looked at like a digital intermediary,” she said. “They’re everywhere, but they’re nowhere.”
Lawmakers are seeking to rein in these chatbots through government regulation. This regulation could also come, in a limited sense, in the form of an initial public offering, like OpenAI is poised to do later this year, according to reports.
Robinson also said that there’s no one singular place to regulate companies since these chatbots are used everywhere, often in a private setting. “That is the basis on which these companies can say, ‘This is private activity and we’re just facilitating it.’”
However, there have been cases in which technology companies are held liable.
In March, a jury found that Meta, which owns Facebook and Youtube, was negligent by making its platforms’ features addictive for children and teens, thereby impacting their mental health. In a separate case, another jury found Meta liable for misleading users and failing to protect children from child exploitation.
The Meta outcome “really surprised us,” Robinson said, because the company did not settle and went to a jury trial.
“Juries are sympathetic to human suffering. Period,” she said.
But these differ from the recent OpenAI lawsuits because they involved children, a protected class.
With the lawsuits accusing OpenAI of causing the criminal or fatal outcomes, Wihbey said that it may be challenging to prove that it was the chatbot and the chatbot alone.
“There may be 20 other pieces of information – like [the perpetrator’s] partner just abandoned them, and they didn’t have a good doctor and they weren’t of sound mind because they had been drinking. So then to say it’s OpenAI is sometimes pretty difficult,” he said. “Right now, it’s a bit of a Wild West situation.”
But the court of law is a good way to test out legal theories and even gain access to company information through the course of the trial, experts said.
“Lawsuits serve both as providing justice to a family that’s grieving because something happened to somebody, but they can also serve this larger function within the policy space of providing us real information” on things like safety measures and how far the chatbots’ guardrails can be pushed, Wihbey said.
Robinson likened the current booming artificial intelligence period to the Industrial Revolution and the rapid advancement of technology of that time, which took several decades to address regulations and safety.
“It’s going to take innovative thinking like it did then about what kinds of organizations these are and how do we reach them,” she said.
Hannah Morse is a news reporter at Northeastern Global News.
© 2026 Northeastern University