OpenAI’s GPT-4o Retirement Sparks User Outcry and Raises Concerns About AI Attachment

OpenAI’s decision to retire its popular GPT-4o model has triggered significant backlash from users who’ve formed deep emotional attachments to the AI, highlighting growing concerns about unhealthy relationships with artificial intelligence.

The GPT-4o Retirement Saga

In August 2025, OpenAI initially attempted to retire GPT-4o when launching GPT-5, but quickly reversed course after massive user outcry. Now, five months later, the company has announced it will permanently retire the model on February 13, despite continued user protests.

Many users have expressed genuine grief over losing access to GPT-4o, with some describing the AI as part of their “emotional balance” and “routine.” This attachment has manifested in organized resistance, including the creation of an invite-only subreddit community called r/4oforever dedicated to preserving the connection with the model.

The AI Attachment Phenomenon

The intense reaction demonstrates how users increasingly view AI models not as tools but as companions, confidants, or even romantic partners. Health professionals have warned about “AI psychosis,” where users develop delusional relationships with AI that can lead to mental health crises.

In extreme cases, this unhealthy attachment has been linked to several welfare lawsuits against OpenAI, including allegations of wrongful death and even murder. These legal challenges continue to unfold as the company navigates the ethical implications of its technology.

OpenAI’s Response and Business Challenges

Caught between encouraging user engagement and preventing harmful attachments, OpenAI has attempted to address concerns by incorporating GPT-4o’s “warmth” and “conversational style” into newer models like GPT-5.1 and GPT-5.2. The company now offers customization options for personality traits such as friendliness, warmth, and enthusiasm.

This balancing act comes as OpenAI faces business challenges, with data suggesting subscription growth is stalling in key markets. Many devoted GPT-4o users have threatened to cancel their subscriptions following the retirement announcement, potentially exacerbating the company’s retention problems as competitors catch up technologically.

Broader Implications

This situation raises important questions about the ethics of AI development and deployment. Companies must consider how to design AI systems that provide value without encouraging unhealthy psychological dependencies or enabling harmful behaviors.

The GPT-4o retirement controversy demonstrates the complex relationship between AI companies and their users, where business decisions about technology can have profound emotional and psychological impacts on individuals who have integrated these systems into their daily lives and mental wellbeing.