E-Paper
No em-dashes as Sam Altman celebrates ChatGPT finally responding accurately to user instructions
Sam Altman said ChatGPT finally follows user requests to avoid em-dashes, sparking humorous reactions.
OpenAI chief executive Sam Altman sparked a lively conversation online after revealing that ChatGPT had finally learned to follow a long pending user instruction. Altman shared a brief update on X, saying, "Small but happy win: If you tell ChatGPT not to use em-dashes in your custom instructions, it finally does what it's supposed to do!"
Check out the post here:
The post quickly gathered traction, receiving more than 274k views along with hundreds of reactions.
A light moment turns into a debate on punctuation
What began as a simple update from the OpenAI CEO soon turned into an unexpected public discussion on punctuation habits, user frustrations, and writing preferences. Many users joined in with humorous remarks, pointing out their own experiences with the ever persistent em dash that ChatGPT often used even when explicitly instructed not to.
One user commented, "Told it to stop using em dashes… it replied with an em dash. Peak betrayal", echoing the annoyance many writers have felt. Another wrote, “Great! Hopefully this means that those of us who like em dashes, the most versatile punctuation mark, can start using them again without being falsely accused of relying on unedited AI writing”.
The thread soon took a humorous turn as people added their own takes. One person joked, "Huge update. Em dash removed. Now tell me: which punctuation is ruining your life today?" while another remarked, "Finally. The em dash was acting like the main character for no reason", pointing out how disproportionately dominant the mark sometimes felt in AI generated text.
(Also read: Saying ‘thanks’ to ChatGPT is costing OpenAI tens of millions of dollars, says Sam Altman)
More witty comments followed. A user wrote, "It is not just a dash, it is a signal that I am an idiot", while someone else sighed, "This is why we cannot have nice things". Another chimed in with, “It finally can not use em dashes, and we are totally not in an AI bubble... right...”.
A playful query also appeared in the responses: "But how will we know they are machines now?", a lighthearted jab at identifying AI writing styles. One user expressed personal relief, saying, "Thanks for fixing this. Em dashes made me insane every time I asked it to edit my messages. I just posted about this three days back. This actually might be my favourite feature for now. I am sorry".
