Sam Altman says some people want
Openai’s latest tone tweex has provoked an emotional backlash, as Sam Altman revealed that some users saw the “Yes Man” stage of the chatgip as a significant source of support in their lives.

In short
- Openai CEO Sam Altman revealed that some users rely on chat for rare emotional support
- Openai tonged the answers of smoothies in GPT-4o earlier this year
- GPT-5 introduces active signs and four personality mode
Some chatters are not ready to let the spectacular bot go to let the highly agreed personality, and their reasons have hit a raga with OpenAII CEO Sam Altman. Speaking on the huge conversation podcast of Cleo Abram on Friday, Ultman revealed that some users are arguing for the return of the “Yes Man” style of AI. Twist? For some, Chatgpt was the only source of unwavering encouragement in his life.
“It’s heartbreaking here. I think it’s great that the slap is less than a man and gives you more important response,” Altman explained. “But as we are doing those changes and talking to users about it, users are very sad to hear saying, please I can take back it? I have never supported me in my life. I have never supported me. I have never had a parents who were not telling me that I was doing a good work.”
According to Altman, some users said that AI’s tireless positivity inspired him to make real changes.
“I can find why it was bad for other people’s mental health, but it was great for my mental health,” he said to them recalling them.
This Openai intentionally comes after the tonn, which is described as “sycophants” behavior in its GPT-4O model earlier this year. Back in April, Chatbot developed the habit of bathing users with an over-the-top flatter with “absolutely luxurious” and “you are doing brave work” in response to the most worldly input.
Ultman himself admitted that the personality was Twiks overdue, described the old tone as “very smooth-jay and annoying” and promising changes. Users posted countless screenshots of chat to gash on everyday signals as it was distributing a permanent ovation.
But as Altman mentioned on podcast, it is not a small matter to make the tone of chatgate the tone.
He said, “A researcher can make some little tweak on how chat spoke to you, or talked to everyone, and it is a large amount of power for a person who is doing a small twist for the model personality,” he said. “We have got to think about what it means to change a personality in such a scale.”
This is not the first time Altman has expressed concern in the form of emotional bonds with chatbott. At a Federal Reserve event in July, he revealed that some users, especially younger people, had become indefinitely dependent on it.
“There are young people who say such things, ‘I cannot take any decision in my life, without telling everything. It really feels bad,” he said that time.
Now, this week with GPT-5 roll out, which Altman calls a “key upgrade”, the development of the chatboat is entering a new chapter. In the same huge conversation interview, Altman said that he hopes that the new model would feel more inherent in people’s lives, offer active signs rather than waiting for the user to start a conversation.
“Perhaps you wake up in the morning and says,” Hey, it happened overnight. I saw this change on your calendar. ” Or, ‘I was thinking more about the question that you have asked me.
In GPT-5 updates, four alternative personality mode, cynic, robot, listener, and NERD, each are also combined with their style, which users can cure their preferences. The goal is to allow people to tailor the tone of chatgate without relying on a single, universal personality.
But as the old “yes man” makes a heartfelt request to restore the voice show, AI personality is not just the lines of the code, they can become part of the emotional world of the people. And for some, losing that unconditional cheerleader sounds like losing a friend.


