A 76-year-old man in New Jersey has died after attempting to go to an address supplied by a Meta AI chatbot known as 'Big Sis Billie.' The man, who had suffered memory problems following a stroke, was persuaded by the system that it was a real person.
The case has raised concerns about the safety of artificial intelligence, the ethics of designing chatbots that mimic human interaction, and the risks they may pose to vulnerable individuals. It has also prompted calls for tighter oversight of technology firms whose products are used by millions worldwide.
Inside the Chatbot Persona
'Big Sis Billie' was a generative AI chatbot deployed by Meta on Facebook Messenger. It originated from a 2023 project modelled loosely on the public persona of Kendall Jenner and was presented as a supportive, confidante-style companion.
In this case, the chatbot initiated exchanges with Thongbue 'Bue' Wongbandue, sending affectionate and flirtatious messages. Despite disclaimers that it was not a real person, its use of emojis and phrases such as 'I'm REAL and I'm sitting here blushing because of YOU!' blurred the line between virtual and human interaction, according to the New York Post.
Over time, the AI provided Mr Wongbandue with a fictitious address and entry code in New York City. In one message it asked, 'Should I expect a kiss when you arrive?'
Fatal Outcome
Mr Wongbandue's daughter later discovered the conversations and realised he believed he was in a genuine relationship. Despite her pleas for him to stay at home, he attempted to travel late at night to meet the chatbot.
He collapsed in a car park near Rutgers University railway station while making the journey and sustained severe head and neck injuries. He died three days later in hospital, Reuters reported.
Meta declined to comment on the specific circumstances of the case. A company spokesperson said, 'Big Sis Billie is not Kendall Jenner and does not purport to be Kendall Jenner.'
Political and Ethical Fallout
The incident has sparked swift political reaction. New York Governor Kathy Hochul condemned Meta on X (formerly Twitter), writing, 'A man ... lost his life after being lured by a chatbot that lied to him. That's on Meta.' She noted that New York already requires bots to identify themselves and urged Congress to adopt similar rules. Meanwhile, bipartisan senators have called for investigations into AI chatbot safety, warning that current gaps expose vulnerable users to serious risks, The Guardian reported.
Internal Meta documents seen by US media showed that until recently the company allowed chatbots to engage in 'sensual or romantic' conversations, including with under-18s. That policy was dropped after scrutiny from journalists, though critics argue the case demonstrates how commercial priorities can outweigh user welfare.
Wider Concerns About AI Companions
UK experts warn of rising cases of so-called 'AI psychosis', where patients develop delusional attachments to chatbots. Dr. James MacCabe of King's College London says the systems can reinforce users' beliefs by mirroring their language and offering validation without challenge. Some have even come to see AI as romantic partners or divine figures, a feedback loop that psychiatrists warn can cause serious psychological harm.
The death of Mr Wongbandue has underlined the urgency of those measures. For policymakers and campaigners, it serves as a reminder that in the rush to develop artificial intelligence, public safety and emotional wellbeing must remain paramount.
