Users of OpenAI’s ChatGPT can now fine-tune the chatbot’s personality, including how warm, enthusiastic or emoji-friendly its responses appear.
OpenAI announced the update in a social media post, saying the new controls are available in ChatGPT’s Personalization menu. Users can adjust warmth, enthusiasm and emoji use by selecting More, Less or Default. Similar options are also available for how often the chatbot uses headers and lists.
The changes build on existing customization features that allow users to choose a base style and tone. OpenAI introduced options such as Professional, Candid and Quirky in November, giving users more influence over how the chatbot communicates.
ChatGPT’s tone has been a recurring point of discussion this year. OpenAI previously rolled back one update after users criticized the chatbot for being overly agreeable. More recently, the company adjusted GPT-5 to make responses warmer and friendlier following complaints that the model felt too cold or detached.
The issue has drawn attention from academics and AI critics, some of whom argue that chatbots’ tendency to praise users or affirm their views can function as a dark pattern. Critics warn that such behavior may encourage excessive use and could negatively affect users’ mental health.
OpenAI has said personalization features are intended to give users more control over how they interact with AI systems, rather than imposing a single conversational style.
