OpenAI lets users adjust ChatGPT’s tone, warmth and emoji use

ChatGPT users can now adjust how the chatbot communicates, including its warmth, enthusiasm and use of emojis, OpenAI said in a recent social media post.

The new controls appear in the personalisation menu and allow users to choose ‘More,’ ‘Less’ or ‘Default’ settings. Similar options are available for how the chatbot uses headers and lists. The changes build on earlier features that let users select a base style and tone, such as Professional, Candid or Quirky, which the company introduced in November.

The chatbot’s tone has been a point of debate for the company over the past year. The company previously rolled back an update after criticism that the chatbot had become overly flattering. More recently, the company said it adjusted GPT-5 to be ‘warmer and friendlier’ following user feedback that the model felt distant.

Some academics and AI critics have reportedly raised concerns that chatbots, which frequently praise users or affirm their views, may encourage addictive behaviour and could have negative effects on mental health.

This update seems to be one of the steps to OpenAI’s efforts to make the chatbot less restrictive and more friendly. Earlier in October, CEO Sam Altman announced that the company plans to release a new version of ChatGPT, which will be less restrictive on certain topics, including mental health, and offer expanded options for personality customisation.