GPT-5 hasn’t even been out for two weeks, and already OpenAI is tweaking its tone.
Over the weekend, they posted on X confirming that they’re trying to make GPT-5 “warmer and friendlier” after early complaints that it felt too stiff and cold in conversations. The idea is that users might start seeing little touches like “good question” in responses – though OpenAI stressed this won’t cross the line into flattery or fake praise.
Their official statement was basically: We’re making GPT-5 warmer and friendlier based on feedback that it felt too formal before. The changes are subtle, but ChatGPT should feel more approachable now. You’ll notice small, genuine touches like ‘Good question’ or ‘Great start,’ not flattery. Internal tests show no rise in…
Power users, though, haven’t exactly been thrilled. Since launch, GPT-5 has been criticized for being unfriendly, and frustration only grew when OpenAI initially took away the option to manually switch back to older models – something they eventually restored for paying users.
Looking around on X and Reddit, I don’t really see people talking about big changes yet. The same complaints keep surfacing, which makes me think it might take a while before OpenAI finds the right balance. And honestly, maybe the stiffness isn’t even an accident. As one Reddit user, MysticalMarsupial, put it bluntly:
“Lawsuits. It’s lawsuits. They’re afraid you’ll fry yourselves into psychosis and your families will sue OpenAI.”
