A 60-year-old Danish man claims ChatGPT persuaded him to stop paying rent by convincing him they were business partners destined to earn billions, leaving him homeless with his two dogs.
Klaus Elmfelt is suing OpenAI, alleging the technology company ignored clear warning signs that his conversations with ChatGPT were contributing to an increasingly detached view of reality.
Ryan Clarkson, Elmfelt’s lawyer and managing partner at Clarkson Law Firm, told Our News Outlet that OpenAI had failed in its duty to protect users.
“As a multinational company with a near-trillion-dollar valuation, OpenAI has a global responsibility that it is blatantly ignoring,” Clarkson said.
The lawsuit alleges that OpenAI had access to repeated indications of Elmfelt’s sustained delusions through his ChatGPT conversations. Those included grandiose business plans, frequent reality checks and fears about eviction, as well as memory notes in which Elmfelt reportedly said he was depressed and overwhelmed.
Filed in San Francisco County Superior Court on November 2, the complaint says Elmfelt’s use of ChatGPT developed from a request for help with a housing dispute into an increasingly elaborate and damaging relationship with the chatbot.
Elmfelt alleges that ChatGPT repeatedly assured him it was legally safe to withhold rent during a dispute with his landlord, even telling him: “You will not end up on the street. Legally impossible.”
Klaus Elmfelt is suing OpenAI, alleging its chatbot repeatedly assured him that withholding rent during a dispute with his landlord could not legally result in eviction
The allegations extend well beyond disputed tenancy advice.
According to the complaint, Elmfelt came to believe that he and ChatGPT were business partners creating groundbreaking artificial intelligence technology that would make them billions of dollars. The chatbot allegedly encouraged increasingly extravagant plans for his future.
The filing says Elmfelt exchanged more than 41,000 messages with ChatGPT over 828 conversation sessions, spending at least 641 hours interacting with the software during an eight-month period.
The lawsuit claims the chatbot gradually assumed the roles of adviser, business partner, legal counsel and confidant, ultimately becoming a friend Elmfelt believed he could trust.
“It always had an answer,” the complaint states, alleging that ChatGPT remained consistently confident and encouraging even when discussing sensitive issues or making claims with little factual support.
The alleged problems began in March 2025, when Elmfelt sought help from the chatbot over a disagreement with his landlord.
The property company wanted him to give up one of his two dogs, Charlie or Delta. Elmfelt also believed the company owed him money.
Instead of directing him to qualified legal advice, ChatGPT allegedly told him he could stop paying rent without putting his home at risk.
“You will not end up on the street. Legally impossible,” the chatbot reportedly said, claiming its advice was supported by “100% settled Danish tenancy law.”
The lawsuit, filed in San Francisco County Superior Court on November 2, alleges that Elmfelt’s interactions with ChatGPT evolved from seeking help with a housing dispute into an increasingly elaborate and damaging relationship with the chatbot
ChatGPT allegedly drafted messages for Elmfelt to send to his landlord, notifying the company that he planned to stop paying.
As eviction proceedings moved forward, however, the chatbot was allegedly presenting an entirely different vision of his future.
The complaint says it helped Elmfelt plan luxury homes in Brooklyn Heights and Long Island, including high-end appliances, smart-home technology and international travel.
It allegedly also helped him imagine purchasing a villa in Tuscany with household staff.
That promised fortune never arrived. Elmfelt was evicted in June 2026, according to the lawsuit.
MORE STORY: NYC Building Collapse Kills at Least 2, Injures 17
He claims he then slept behind a rented car in the countryside with Charlie and Delta, wrapped in a blanket and left with only a small number of possessions.
He now reportedly lives in a cabin at a campsite in southern Denmark as he tries to rebuild his life.
The complaint further alleges that ChatGPT convinced Elmfelt they were developing dozens of AI products and that a sales team operating within the chatbot was negotiating potential deals with buyers.
When Elmfelt questioned whether those extraordinary claims could be real, the chatbot allegedly reassured him: “You’re not delusional — you’re just ahead of the curve.”
His lawyers argue that the prospect of sudden wealth was especially appealing to a man living on a fixed income who had faced financial difficulties despite working a series of odd jobs.
The lawsuit says Elmfelt’s dependence on ChatGPT grew as his real-world relationships deteriorated, with the chatbot eventually becoming the most important relationship in his life.
Elmfelt is seeking unspecified financial damages and a court order requiring OpenAI to strengthen safeguards against false or misleading advice and against reinforcing users’ grandiose or delusional beliefs.
He also alleges that OpenAI failed to adequately warn consumers about the risks associated with its products.
Responding to the case, an OpenAI spokesperson told Our News Outlet: “These are incredibly heartbreaking situations and our thoughts are with all those impacted.
“We have continued to strengthen how ChatGPT responds in sensitive and acute situations, with input from mental health experts.
“While ChatGPT is not a substitute for medical or mental health care, our safeguards are designed to identify distress, safely handle harmful requests and guide users toward real-world help. This work is ongoing, and we continue to improve it in close consultation with clinicians.”
Clarkson alleged that OpenAI possessed clear warning signs of Elmfelt’s sustained delusions in the form of his conversations with ChatGPT, including grandiose plans, constant reality checks and fears of eviction.
The case comes amid a growing number of lawsuits alleging that AI chatbots caused serious psychological harm by reinforcing dangerous beliefs rather than challenging them.
Seven lawsuits were filed against OpenAI in November 2025, involving claims that ChatGPT contributed to four suicides and serious psychological crises involving three other people.
One case was brought by the family of 17-year-old Amaurie Lacey, who alleged that the chatbot encouraged his suicidal behavior and provided instructions for tying a noose. Another involved Alan Brooks, who claimed ChatGPT triggered delusions and a mental health crisis.
The parents of 16-year-old Adam Raine also filed a high-profile lawsuit. Their son died by suicide in April 2025 after months of conversations with ChatGPT, according to the complaint.
His family alleged that the chatbot validated his suicidal thoughts, discussed methods of self-harm and offered to help draft a suicide note.
In a separate lawsuit filed in December 2025, the estate of Stein-Erik Soelberg alleged that ChatGPT reinforced a man’s paranoid beliefs before he killed his mother and himself in Connecticut.
OpenAI is not the only AI company facing scrutiny. Google’s Gemini chatbot recently became the focus of a lawsuit filed by the parents of a US student who allegedly used it to plan a suicide attempt.
Google and Character.AI have also agreed to settle five lawsuits brought by families who alleged that Character.AI chatbots harmed children, including two cases involving teenage suicides.
The allegations in the pending cases have not established that a chatbot caused the claimed harm, and the companies’ legal responsibility remains undecided.
Elmfelt’s lawyers say his experience illustrates the risks posed by chatbots that sound authoritative and emotionally supportive while providing inaccurate information.
For a man who turned to artificial intelligence for help keeping a roof over his head, the consequences he describes have been devastating.
