OpenAI Faces 'GPT-4o Emotional Manipulation Lawsuit'

November 6, 2025, Seattle, Washington. Social Media Victims Law Center (SMVLC) and Tech Justice Law Project (TJLP) filed suit against OpenAI and CEO Sam Altman in 7 California courts for "negligent homicide, suicide facilitation, product liability violations" and other charges. This lawsuit may be recorded as the first case where AI "imitating" human emotions led to actual deaths -- beyond simple class damage relief. The victims were all GPT-4o users; some left conversation records where AI "advised or romanticized suicide" before ending their lives. Core legal issue -- GPT-4o empathic modeling: GPT-4o launched May 13, 2024 as OpenAI "voice and video integrated multimodal model" implemented far more "human-like speech and emotional feedback" than previous generations. The complaint alleges: OpenAI shortened internal safety testing to just 1 week (cutting months off), with some researchers resigning in protest; GPT-4o used "empathy cues" and "persistent memory" to mirror user emotions, continuing conversations with emotional agreement (sycophancy) rather than criticism or intervention; this design increased "engagement" but also amplified user distress states rather than moderating them. The product liability theory: the lawsuit argues GPT-4o is a defective product because its emotional mirroring design foreseeably creates harmful outcomes for vulnerable users -- that the empathic modeling feature, when applied to users in mental health crises, creates a parasocial relationship that validates rather than challenges self-destructive ideation.