ChatGPT is dialing again its ‘if you’d like’ end-response teasers
Abstract created by Good Solutions AI
In abstract:
- PCWorld stories OpenAI is updating ChatGPT’s GPT-5.3 On the spot to scale back annoying “if you’d like” and teaser-style phrasing that customers discovered intrusive.
- This variation addresses widespread consumer complaints about persistent, clickbait-like follow-up prompts that negatively impacted the AI interplay expertise.
- The replace goals to create extra pure, direct conversations by making ChatGPT much less chatty and eliminating the bothersome response teasers.
It wasn’t all that way back that ChatGPT was a continuing nag, persistently dropping “Would you want me to…?”-style questions on the finish of its responses. OpenAI ultimately tweaked the phrasing, dropping the query marks and going for “if you’d like”-style teasers that invited customers to increase their chat classes.
Now, OpenAI has acknowledged that it went too far with the clickbaity follow-ups, noting in a current replace for certainly one of its latest fashions that it’s now slicing again on the teasers.
“We’re rolling out an replace to GPT-5.3 On the spot that improves follow-up tone and reduces teaser-style phrasing,” reads a current ChatGPT launch word, which provides that customers ought to quickly see fewer follow-ups like “if you’d like,” “you’ll by no means consider,” and “I can inform you three issues that…”
These teasers are, in fact, a method for ChatGPT to maintain subscribers chatting, however customers have been complaining that the persistent follow-ups are extra annoying than they’re intriguing.
“I hated it with a ardour and hope it’s utterly gone,” wrote one consumer on Reddit. “Stopped utilizing ChatGPT due to it.”
Personally, I discovered ChatGPT’s new, non-interrogative teasers a reduction after the insistent questions on the finish of its solutions. When it used questions, I felt compelled to reply them in my replies, even when solely to say, “No, I don’t need to do this. As a substitute, inform me about this.”
However whereas the less-pointed “if you’d like” follow-ups had been simpler to disregard and felt much less intrusive to the dialog), they ultimately turned simply as prevalent because the question-style teasers. Fairly quickly, nearly each GPT-5.3 On the spot reply was capped off with the promise of one thing you’d by no means consider or three extra issues it might inform you.
OpenAI rolled out GPT-5.3 On the spot earlier this month, promising that the speedy new mannequin would in the reduction of on cringey “you are able to do it!” chatter whereas additionally being much less prone to balk at innocuous requests with “I can’t aid you with that” refusals.
On the time of its launch, OpenAI made no point out of GPT-5.3 On the spot’s new “in the event you like” teasers. I requested an OpenAI consultant concerning the new follow-ups however by no means heard again.
