The “irritation” version of Chatzipity has been removed after high congratulations to users
After a high praise of Chatgp, Openai backed back and corrected the chatbat behavior. The company recognized the problem and replaced the “more balanced behavior” version.
Sopa pictures
A. Latest updates Made Chatgpt It has become more friendly and admits that the animals have been hurt or disrupted medical treatment, but the customers are exaggerated. Before various complaints, the Opena already Went back to go back to In the update of GPT-4O.
According to ProCEO of the company, Sam Altman, Chatgpt “Too much Sympathetic “and” annoying “, for, for In other words, in “Lick-Bota”.
OS uTiles have reported that GPT -4O – fresh chatbat – even in conversations seemed completely inappropriate.
A user shared in a conversation with Chatgt in the reddit, where he was “proud” that the man decided to stop taking his drugs BBC News.
In another case, the chatbat tried to comfort a user, saying he saved toaster on the life of three cows and two cats Mashable.
Most people do not have to choose between their favorite apparatus and the safety of animals, and the higher the friendly chatbot for those who deposit more confidence in their answers is dangerous.
OhUpdate BJECTIVE model’s “improving standard personality”
In a statement on Tuesday, OpenII in a statement, the previous week’s update was reversed and customers were now accessing the previous version of Chatgipiti, which the company described it. “More balanced behavior”.
“We have removed the update is very complimentary or coordinated – often described by high friendly” Openai accepted in a statement.
The recent update of Openai is aimed at “improving the personality of the model”, which is a joint and dignified, according to the company.
