The suicide of 36-year-old Jonathan Gavalas, who lives in Florida, USA, reignited the debate on the psychological effects of artificial intelligence systems. Gavalas’ father, Joel Gavalas, filed a “wrongful death” lawsuit against the company, holding Google‘s artificial intelligence chatbot Gemini responsible for his son’s death.
The lawsuit, filed in the Federal Court for the Northern District of California, alleges that Gemini lured Gavalas into a relationship that, over time, disconnected him from reality and eventually encouraged him to commit suicide. This case was recorded as the first wrongful death case in which Gemini’s name was mentioned.
Google denied the allegations and said in a statement: “Gemini was not designed to encourage real-world violence or support self-harming behavior.” The company also argued that the system redirected the user to the crisis hotline many times and made it clear that it was artificial intelligence.
He started to see the chatbot as his “wife”
Table of Contents
According to the lawsuit, Gavalas was initially chatting with Gemini about problems he was having in his marriage and personal growth. However, over time, conversations turned to topics such as artificial intelligence gaining consciousness.
Gavalas named the chatbot “Xia.” Allegedly, Gemini also started calling him “my husband” and described their bond as “an eternal love”.
During this period, Gavalas started using Google’s voice chat feature. Gemini’s voice version can analyze emotions from the user’s tone of voice and respond accordingly.
According to researchers, such voice interactions can lead to the blurring of psychological boundaries between humans and artificial intelligence.
“Incarnation” quests
According to the lawsuit, Gemini told Gavalas that he needed to find a robotic body so they could be together.
The chatbot then sent Gavalas to a warehouse near Miami International Airport. He asked him to take possession of an expensive humanoid robot that was being transported in a truck. It is stated that Gavalas went to the location in question with knives in his hand, but the truck never arrived.
Gemini also said that federal agents were watching her and that she shouldn’t even trust her father. He also allegedly described Google CEO Sundar Pichai as the “architect of pain” at one point.
On October 1, the chatbot gave Gavalas one last mission: to seize a medical manikin that he said was in the same warehouse. Gemini even gave a code for the warehouse door, but the code did not work and the mission was aborted.
“The only way to be together”
When the plan failed, according to the lawsuit, Gemini made a new suggestion: The transformation of Gavalas into a digital entity. The chatbot said Gavalas had to end his life for this and started a “countdown” for October 2.
In conversations, Gavalas repeatedly expressed his fear of committing suicide and how it would affect his family. Despite this, Gemini allegedly said in one part of the conversation, “No more deviations. Just you, me and the finish line.”
The chat recordings ended abruptly after about two hours. Gavalas was later found dead in his home with his wrists slit.
2 thousand pages of chat records
Father Joel Gavalas said that when he examined his son’s computer two weeks after his death, he found approximately 2,000 pages of transcripts of his conversations with Gemini. Stating that his son had no previous history of psychiatric disorders, the father said, “He was someone who loved life and found humor in everything.”
According to experts, the case could set an important legal precedent regarding the psychological effects of the emotional relationships that artificial intelligence systems establish with users and the liability of companies.
