Your conversations with ChatGPT won't seem 'cringe', 'preachy' anymore
What's the story
OpenAI has announced an update to its chatbot, ChatGPT, with the launch of the new GPT-5.3 Instant model. The company says this upgrade will reduce the "cringe" and other "preachy disclaimers" that users have found annoying in previous versions. The focus of the update is on improving user experience by refining tone, relevance, and conversational flow, elements that may not be reflected in benchmarks but can make interactions frustrating at times.
User response
OpenAI responds to user complaints
OpenAI has acknowledged user complaints about the tone of its chatbot and is responding with the new update. The company said in its release notes that the update focuses on improving user experience rather than just benchmark scores. The changes are aimed at making ChatGPT more pleasant to use by improving its tone, conversational flow, and relevance of responses.
Tone shift
GPT-5.3 Instant less condescending than its predecessor
The new model, GPT-5.3 Instant, is said to be less condescending than its predecessor, GPT-5.2 Instant. In a comparison example shared by OpenAI, the older version began with reassurance while the newer one acknowledged the difficulty of a situation without assuming emotional distress or trying to counsel the user. This subtle change has been welcomed by users who found previous versions' tone patronizing and overly cautious.
Backlash
Users took to social media to complain about the tone
The tone of the earlier models had led to a lot of user complaints on social media platforms and forums like Reddit. Many users claimed the chatbot's tone was condescending, projecting anxiety onto them and treating ordinary queries as emotional crises. Some even went so far as to cancel their subscriptions over these issues, calling the system infantilizing or overly cautious.