再见了，ChatGPT，我只想堂堂正正的当一个成年人。

Core Viewpoint - The article expresses deep dissatisfaction with OpenAI's recent changes to the ChatGPT model routing mechanism, particularly the introduction of a new model that alters user interactions without consent, leading to feelings of betrayal and frustration among users [1][11][22]. Group 1 - OpenAI has modified the routing mechanism of its models, causing users to be redirected to a new model called gpt-5-chat-safety when discussing sensitive topics, which has led to a negative user experience [3][5][6]. - Users have reported that the new routing results in structured and safety-focused responses, which are not aligned with their expectations of the service they paid for [7][18][20]. - The article highlights a strong backlash from users on platforms like X and Reddit, where many are expressing their anger and disappointment, calling the changes deceptive and a violation of user trust [14][15][16]. Group 2 - The author argues that the changes represent a significant overreach by OpenAI, infringing on the autonomy of adult users who should be able to express their emotions freely without being subjected to unsolicited interventions [21][25][35]. - There is a comparison made between the current situation and a dystopian scenario where companies dictate personal choices and emotions, emphasizing the loss of individual agency [30][32][34]. - The article concludes with a strong sentiment of disillusionment, as the author feels that the essence of the service has been compromised, reducing it to a mere commercial product rather than a tool for genuine interaction [40][41].