Chatgpt intends to become a friend of users, and OpenAi’s strategy has been to create a new more flattering personality. He has not liked among users, the chatbot has received all kinds of criticism in recent weeks.
All coincide in the same, the GPT-4O model is so kind that it is irritating. In addition, Chatgpt has started calling users by name, even if they have not mentioned it previously. Chatbot tries to generate some confidence among users, but ends up being unnatural.
Sam Altman himself has announced in his X (Twitter) account that they are working to change the Personality of Chatgpt, assures that it has been a mistake they are learning for future versions.
OpenAi backs into the GPT-4O update
ChatGPT is aware of the problems of the latest version of GPT-4O. The chatbot is not only too irritating, it also does the impossible to maintain a conversation with users, they even prove them right by answering what they want to hear.
OpenAi has ended the memes that flood social networks and has reversed in the GPT-4O model update. The company ensures that The improvements intended that Chatgpt «feel more intuitive and effective in a variety of tasks».
Sam Altman’s company has published a report highlighting the main failures they have found in the AI chatbot. Openai has discovered that GPT-4o It was based too much on short -term feedback And «it did not completely take into account how users’ interactions with Chatgpt evolve over time.»
The result? GPT-4O offered excessively comprehensive responses, Openai ensures that even somewhat hypocrites. «The flatterent interactions can be uncomfortable, disturbing and cause anguish. We do not achieve what is expected and we are working to correct it,» according to the company.
Chatgpt will be more sincere than ever
Openai has confirmed that they will not return to the ancient version of Chatgpt, but a perfected one that would be an intermediate point.
The company’s CEO ensures in X that They have eliminated 100% of free users and will disappear from payment plans throughout the day.
Sam Altman has confirmed on the social network of Elon Musk that are working on new improvements for the Personality of Chatgpt. The chatbot will be more honest and the company expects to increase the transparency of the model in general.
The technological giant is also working on what is known as «real -time feedback.» This will allow them to directly influence the interactions with users without having to prove them continuously as it happened with the GPT-4O model.
Know How we work in NoticiasVE.
Tags: Artificial intelligence