GPT-5 has been on the market for less than fourteen days, and OpenAI is already modifying its approach. In a message on X over the weekend, OpenAI revealed that it is making GPT-5 “warmer and friendlier” in response to initial impressions claiming it was overly formal and aloof during user interactions. Respondents may notice expressions like “great question” in the replies, though OpenAI made it clear that this won’t involve undue flattery.

Frequent users of ChatGPT have faced a variety of problems with the new model since its release earlier this month, with its supposed lack of friendliness often highlighted on social platforms. Additionally, it created challenges for Sam Altman and OpenAI that the ability to revert to previous models was eliminated, although it was later reinstated for paying users.

Investigations on sites such as X and Reddit suggest there has not been a significant shift in how individuals are discussing GPT-5 as of yet. Many of the same grievances continue, indicating that it might take some time for OpenAI to comprehensively tackle the matter. As Reddit user MysticalMarsupial perceptively pointed out, OpenAI might prefer GPT-5 to maintain a degree of emotional neutrality for a sensible reason.

“Lawsuits. It’s lawsuits,” the user remarked. “They are worried that users could face psychological challenges, resulting in potential lawsuits against OpenAI.”