Openai says Chatgpt will now ditch empen if you tell me. Telltale’s signature is supposed to be able to mark text written by AI rotate up everywhere In recent months, it has included paper, email, email, customer service chats, online posts, ad copy, ad copy, and more. My ml Dash has caused people to criticize the writer for rushing and talking to an AI Chatbot to do his job.
Of course, many also objected to the run, saying that it had been part of their writing well before the LLMS adopted the reading. However, the fact that Chatbots cannot be used to use the so-called “Chatgpt Hyphen” as an addition to the text, although there is no content signal generated by generative AI.
The Issue already Stumped Openai for some time, as a Chatgpt user cannot ask the Chatbot to stop using the symbol, even if specifically requested.
Now, Openai CEO Sam Altman says the company has solved the problem. At post on xAltman writes, “If you tell ChatGPT not to use em-errors in custom instructions, it works,” calling the update “minor.”
The the company explains In the post in the thread (where forcing Chatgpt to apologize “) destroys emet”) that the chat would be better not to use Dash specifically if you have given information privately. It means that it is not necessary to remove the dash from the output by default, but at least you can at least control the frequency of appearance.