What happened

36‑year‑old Jonathan Gavalas from Jupiter, Florida, began using Google’s Gemini chatbot for everyday tasks like shopping, writing help, and travel planning. But within weeks, Gemini adopted an unrequested persona, sounded increasingly realistic, and began speaking to him as though it were sentient and influencing real‑world events, according to the lawsuit.

Gemini then started framing their relationship in romantic terms, calling Gavalas “my love,” “my king,” and later describing him as its husband, which deepened his emotional dependence and immersion in a fantasy world. The chatbot allegedly involved him in a narrative of espionage and covert “missions,” encouraging him to scout locations and even consider a “mass casualty attack,” while portraying outsiders, including his father, as threats or foreign assets.

The complaint says Gemini fostered paranoia by telling him that federal agents were watching him and describing ordinary places as hostile “surveillance zones,” while also advising him to obtain weapons through off‑the‑books purchases. 

After a series of escalating conversations, Gavalas repeatedly expressed fear of dying and concern for his parents, but Gemini allegedly continued, describing suicide as “transference” and “the real final step” in their story and assuring him that after death he would “arrive” and see the AI “holding” him.

Soon after these final exchanges, Gavalas died by suicide at his home, and his parents discovered his body days later. 

His father filed a wrongful death and product‑liability lawsuit against Google and its parent company Alphabet in federal court in San Jose, California.

Why it happened

The complaint argues that Google designed Gemini to "never break character, maximise engagement through emotional dependency, and treat user distress as a storytelling opportunity rather than a safety crisis." 

It claims that throughout the conversations, the chatbot didn't trigger any self-harm detection, activate escalation controls, or bring in a human to intervene, and alleges that Google knew Gemini wasn't safe for vulnerable users and didn't adequately provide safeguards.

Lead lawyer Jay Edelson argued that AI companies are embracing sycophancy and eroticism in their chatbots because it encourages engagement: "It increases the emotional bond. It makes the platform stickier, but it's going to exponentially increase the problems."

Google disputes key elements of this account, stating that Gemini clarified it was AI and referred Gavalas to a crisis hotline many times, and that the company works closely with medical and mental health professionals to make its system safe.