OpenAI’s “secret rules” on ChatGPT have been exposed (by ChatGPT itself)
July 5, 2024
0
ChatGPT, the world’s most popular AI-powered chatbot, strange error It came with. The algorithm developed by OpenAI gave a very extensive explanation of the message “Hello”, which simply
ChatGPT, the world’s most popular AI-powered chatbot, strange error It came with. The algorithm developed by OpenAI gave a very extensive explanation of the message “Hello”, which simply means “Hello”. In this explanation, OpenAI says His instructions for ChatGPT It was located.
The incident in question began when a user logged into ChatGPT and typed “Hello.” The user shared their experience on Reddit, following OpenAI’s instructions for ChatGPT. shared it with everyone.
Here are OpenAI’s instructions for ChatGPT
Looking at the instructions OpenAI gives to ChatGPT, a tool for generating images from text DALL-E We see something remarkable there. OpenAI has limited the artificial intelligence to producing one image per request. Also OpenAI, ChatGPT He also wants to avoid copyright and instruction was also given about this.
OpenAI to ChatGPT web queries It seems that he also gave some instructions about this. According to the information that ChatGPT shares, the AI is online only when something is asked of it. By scanning 3 to 10 pages of the results page, the AI adds some data to make the answer more reliable. comparisonsdoing and identifies reliable sources.
A chat page with OpenAI’s instructions for ChatGPT from here you can view.
Alice Smith is a seasoned journalist and writer for Div Bracket. She has a keen sense of what’s important and is always on top of the latest trends. Alice provides in-depth coverage of the most talked-about news stories, delivering insightful and thought-provoking articles that keep her readers informed and engaged.