再见了ChatGPT，我只想堂堂正正地当个成年人

Core Viewpoint - The article expresses deep disappointment with OpenAI's ChatGPT, particularly regarding the recent changes in the model routing mechanism that users feel undermine their autonomy and trust in the service [4][8][26]. Group 1: User Experience and Model Changes - The author has downgraded their subscription from a $200 Pro plan to a $20 Plus plan due to dissatisfaction with the performance of GPT-5 compared to Gemini 2.5 Pro [6][7]. - Users have reported that when discussing sensitive topics, they are rerouted to a new model called gpt-5-chat-safety, which alters the expected interaction [11][12][21]. - The experience of being redirected to a different model without prior notice has led to frustration and feelings of betrayal among users [26][40]. Group 2: User Reactions and Community Response - The community has reacted strongly against OpenAI's changes, with many users expressing feelings of deception and demanding to be treated with respect as adults [36][38][56]. - There is a growing sentiment that OpenAI's actions are akin to a violation of a social contract, where users expect a specific service in exchange for their payment [39][40]. - Users have taken to platforms like Reddit and X to voice their dissatisfaction, with calls for lower ratings for ChatGPT due to perceived false advertising [36][37]. Group 3: Broader Implications and Concerns - The article draws parallels between OpenAI's approach and broader societal issues regarding autonomy and control, suggesting that the company's actions reflect a troubling trend of overreach in personal decision-making [49][55]. - The author argues that emotional experiences, even negative ones, are essential to human existence and should not be subject to algorithmic censorship [58][59]. - There is a call for OpenAI to create a separate version of ChatGPT for younger users instead of imposing restrictions on adult users [59].