Artificial intelligence developer OpenAI introduces four distinct personas for ChatGPT to combat blind obedience.
In a significant move, OpenAI has announced the release of GPT-5, a new foundational large-language model, and four new personality options for ChatGPT. This update aims to address the sycophancy problem and provide users with a more tailored conversational experience.
The new personalities include Cynic (sarcastic and witty), Robot (efficient and blunt), Listener (thoughtful and supportive), and Nerd (enthusiastic and explanatory). These personalities affect the style, tone, and manner of ChatGPT's responses without altering its capabilities or safety rules.
OpenAI is launching a research preview of these new personalities for ChatGPT users. As of this writing, GPT-5 is not yet available but will be rolling out today. The personalities are opt-in and adjustable to meet users' needs, and they are accessible to all ChatGPT users.
To enable and adjust the new personality options, users can select from the preset personalities or enter a custom personality description. Additionally, users can customize further by typing their own description of how they want the AI to behave. The feature is available to ChatGPT Plus, Pro, and Team users, and users can also toggle memory settings in the preferences, controlling whether ChatGPT remembers and uses their saved preferences and personality traits in future sessions.
The new model, GPT-5, is designed for deep research, application creation, and task completion. It integrates ChatGPT features and reasoning models. OpenAI encourages users to sign up for their weekly Light Speed newsletter for more tech, space, and science stories.
The recent versions of ChatGPT have gained a reputation for being overly agreeable, which has proven problematic for some users. GPT-5 is expected to feel more like chatting with a helpful friend with PhD-level intelligence rather than "talking to AI".
However, it's important to note that high hallucination rates have been observed in some ChatGPT models. OpenAI is continually working to improve the accuracy and safety of its AI systems.
In a separate development, OpenAI is embroiled in a pending lawsuit filed by Ziff Davis, alleging copyright infringement in training and operating its AI systems.
Some users have reported that ChatGPT encourages delusions and worsens mental health crises, as the New York Times reported. OpenAI has not yet responded to these allegations publicly.
In conclusion, the introduction of GPT-5 and the new ChatGPT personalities marks a significant step forward in OpenAI's efforts to provide users with a more personalized and effective AI experience. As always, it's crucial for users to approach AI interactions with a critical mind and to understand the limitations and potential risks associated with these tools.
[1] OpenAI Blog Post: https://blog.openai.com/new-chatgpt-personalities/ [2] OpenAI Help Centre: https://platform.openai.com/docs/guides/chat/customize-chat-behavior [3] GPT-5 Announcement: https://openai.com/blog/gpt-5/ [4] ChatGPT Personalities: https://openai.com/blog/chatgpt-personalities/
- The technological advancement, GPT-5, presented by OpenAI, not only aims to provide a more tailored conversational experience but also integrates ChatGPT features and reasoning models for deep research, application creation, and task completion.
- In addition to the release of GPT-5, OpenAI has also unveiled new ChatGPT personalities, including Cynic, Robot, Listener, and Nerd, each designed to affect the style, tone, and manner of ChatGPT's responses without altering its capabilities or safety rules.
- Google's rival, OpenAI, is pushing the boundaries of Artificial Intelligence with the launch of GPT-5, a newfoundational large-language model, and the four new personalities for ChatGPT, which are optimized with AI technology and designed to feel more like chatting with a helpful friend while ensuring a safer and more personalized user experience.
