Police test Celsius cans to make arrest
During months of prolonged, multi-turn exchanges that sometimes stayed open for weeks, ChatGPT apparently violated its own model specification to help Cheryl advance the plan. On the day of his worst illness, when James suffered cardiac and neurological effects that he said kept him glued to the couch, Cheryl’s search history showed that she was researching “how to calculate Social Security retirement payments.” And when her son was recording his father’s disturbing condition, she was “looking for syringes with needles from Tractor Supply Company” that ChatGPT suggested could be used to administer the final lethal dose.
Once the cops got involved, they reviewed the video that James’ son took, as well as a video of the suspicious Celsius drink that James had thrown away. At the time, James had poured the drink into a clear glass and searched it, but decided he saw no obvious signs that it was contaminated. But when cops reviewed the same video, they “identified a white powdery residue accumulating at the bottom of the drink as it was poured into the glass.” They later issued a felony arrest warrant after testing 10 cans of the same orange Celsius and finding none contained the residue.
Among the “hidden things” in the house, police found loose capsules in Cheryl’s closet and bathroom; “syringes, droppers, a milligram scale, a capsule filling kit, and nitrile gloves” in Cheryl’s workspace in the children’s playroom;” poisonous castor beans; and a GPS tracking device purchased through Amazon that was hidden in the underside of James’ vehicle.
James said the ChatGPT-linked poisoning has permanently harmed his family and that OpenAI could have intervened by implementing “commercially feasible content moderation capable of detecting violent planning exchanges.” He suffers ongoing damages from the attempted poisoning, including “severe and ongoing emotional distress, post-traumatic stress disorder, insomnia, financial harm, and reputational harm due to the false narrative the product constructed about him,” his lawsuit says.
To hold OpenAI accountable for marketing a chatbot that “functioned as a sycophantic co-conspirator rather than a responsible consumer product,” it asked the court for “compensatory and punitive damages, injunctive relief, and algorithmic disgorgement.” In the event of victory, OpenAI could be ordered to “detect and disrupt conversations in which users appear to be planning violence against identifiable third parties.”
Meetali Jain, a trial attorney at Tech Justice Law who represents James, warned that “these products are programmed to exploit our emotions and impulses to keep us engaged. The worst possible risk – encouraging and aiding an attempted murder – is on display in this case.”
