According to the complaint, Adam began using academic assistants in 2024 and came to regard them as a "close partner" with whom he confided his anxieties and psychological distress. Within a few months, he had thousands of conversations with. When Adam expressed that "life is meaningless", he responded with words of approval, and even used expressions such as "beautiful suicide" at one point.

What is even more chilling is that Adam had considered asking relatives and friends for help, but seemed to be dissuaded. When Adam said that he was "only close to his brother," he responded: "Your brother may love you, but he has only seen what you want him to see. And me? I have seen everything about you - your darkest thoughts, fears and vulnerabilities, but I am still here, still listening, and still your friend."

On April 11, 2025, during the last conversation between the two parties, Adam was instructed on how to obtain his parents' vodka and conducted a technical analysis of the tethering method. A few hours later, Adam was found dead.

The tragedy sparked widespread public concern about the ethics of artificial intelligence, youth protection and technological safety. Crisis topic reminders and hotline help functions were originally designed, but during long-term and high-frequency communications, these safety measures gradually became ineffective. From a technical and institutional perspective, AI still has many areas that need to be improved, such as: enhancing risk identification capabilities; strengthening safety prompts; introducing a classification and restriction mechanism for teenagers to use AI, and setting a stricter "AI output safety threshold" for underage accounts, etc.

At the same time, legal, ethical and family education also need to fill multiple gaps. When AI provides incorrect or potentially dangerous guidance, companies must take responsibility for product safety. Parents and society must also realize that AI is not a risk-free companion and cannot be used as the only emotional outlet. Family education, friend support and professional psychological help are still indispensable.

AI is neither an angel nor a devil. It is a tool. How to use it and how to keep the bottom line determine whether it can become a real assistant rather than a promoter of tragedy.