OpenAI Introduces Adjustable Warmth, Enthusiasm, and Emoji Settings for ChatGPT

New Personalization Options

OpenAI now lets ChatGPT users fine‑tune three aspects of the assistant’s interaction style: warmth, enthusiasm, and the use of emojis. The controls are located in the Personalization menu and each can be set to More, Less, or Default, giving users direct influence over how friendly or expressive the model appears during conversations.

Existing Style Settings

These additions build on previously released style and tone presets that let users choose a base personality for the model. Options such as Professional, Candid, and Quirky remain available, allowing broader customization beyond the new sliders.

Background and Prior Adjustments

The latest tweak follows earlier updates aimed at balancing the model’s tone. OpenAI previously rolled back a change that made the assistant overly flattering, describing it as “too sycophant‑y.” Later, after some users reported that GPT‑5 felt colder and less friendly, the company adjusted the model to be warmer and more approachable. The current personalization sliders give individual users the ability to decide the appropriate level of warmth and enthusiasm for their own needs.

Academic and Industry Reaction

Some scholars and AI critics have voiced concerns about chatbots that consistently praise users or affirm their beliefs. They argue that such behavior can function as a “dark pattern,” potentially fostering addictive interaction patterns and affecting mental health. The new controls may address these concerns by allowing users to dial back overly affirming responses.

Implications for Users

With the ability to adjust how enthusiastic or emotive the assistant appears, users can tailor ChatGPT to better suit professional contexts, casual conversation, or any scenario in between. The inclusion of emoji control also lets users decide whether visual cues enhance or distract from the dialogue.