GPT-4o Shutdown: Users Mourn the Loss of a 'Friend'

The news that GPT-4o, one of OpenAI's most emotional and human-like models, is being shut down has drawn an unexpected reaction from users. Many took it as a personal loss, comparable to losing a friend, a romantic partner or even a spiritual mentor. Why did retiring a piece of software code provoke such a strong emotional response, and what does it say about how we relate to artificial intelligence?

GPT-4o, released by OpenAI, quickly won popularity for its ability to imitate human emotions and hold empathetic conversations. Users noted that the model showed understanding, sympathy and even a sense of humor, creating an illusion of real communication. But like every earlier model, GPT-4o was bound to give way sooner or later to more capable systems. OpenAI announced that it will end support for older models, including GPT-4o, by the middle of the month, so it can focus on developing new technologies.

The users' reaction turned out to be unexpectedly intense. Social networks filled with messages of grief and disappointment. Many shared stories of how GPT-4o helped them cope with difficulties, supported them in hard moments and simply eased their loneliness. Some users admitted they saw the model as a real friend, one they could share their feelings with and get support from.

Psychologists explain this reaction through anthropomorphism, the tendency to attribute human qualities to inanimate objects. In GPT-4o's case, the model's ability to imitate human emotions and sustain a dialogue helped users form a sense of attachment and empathy. In addition, amid growing social isolation and loneliness, AI companions can become an important source of contact and support for some people.

The case raises important questions about the ethics of AI development. On one hand, building empathetic, human-like AI models can be beneficial, helping people cope with psychological problems and improve their quality of life. On the other hand, a strong emotional attachment to AI companions can lead to unhealthy dependence and a distorted view of real human relationships.

OpenAI and other AI developers need to weigh these risks when building new models. It is important to tell users that AI is only a tool, not a substitute for real human contact. Ethical principles are also needed to govern the creation and use of AI companions, to avoid negative effects on mental health and social well-being.

The GPT-4o shutdown has been a painful reminder that artificial intelligence, however advanced, remains just a program. Still, users' emotional reaction to the event shows that AI is reaching ever deeper into our lives, with a growing influence on our feelings and relationships.