OpenAI announced late Friday that its newest model, GPT-5, is receiving an update designed to make it “warmer and friendlier.”
The change follows feedback from users who said they preferred the personality of GPT-4o, the earlier version. CEO Sam Altman admitted the GPT-5 launch was “a little more bumpy than we’d hoped for,” as some found the new model too direct.
What’s Changing
The company said the adjustment is subtle, meant to make GPT-5 feel “more approachable.” Examples include short touches like, “Good question” or “Great start,” which OpenAI insists are not flattery but ways to make the model feel more natural.
In a social media post, OpenAI noted that internal testing showed “no rise in sycophancy compared to the previous GPT-5 personality.”
At a dinner with journalists earlier this week, executives tried to steer the conversation toward plans beyond GPT-5. Still, the launch’s rocky rollout dominated the discussion. VP Nick Turley described GPT-5 as “very to the point,” but said the new update would make the model feel “warmer,” echoing the announcement made Friday.
What The Author Thinks
This update might sound small, but it highlights something bigger: people want AI to feel less robotic and more like a companion. Even if accuracy and speed matter most, the tone affects trust. A friendlier GPT-5 may help people feel more comfortable using it every day, especially as AI becomes part of conversations, work, and personal life.
Featured image credit: MrPaloma
For more stories like it, click the +Follow button at the top of this page to follow us.