ChatGPT, will you please stop asking me questions?

“Do you want me to regenerate the document with shorter paragraphs and subheads for better readability?" ChatGPT kept asking me after I prompted it to do just that. Seconds later, it would come back with the same errors, confidently telling me it was now perfect—except it wasn’t.

I would show it an image of a page, and it would be very sorry. It would describe in detail what went wrong and how to fix it. Stuck in some loop, it would keep repeating this routine.

I was on the point of giving up when it finally gave me a usable file. But ChatGPT still had an appetite for more. “Would you like me to take your report and re-run it through this exact process from the top (treating it as a raw PDF to DOCX conversion first) so you finally have the cleanest possible version?"

I told it to stop asking me questions and go to sleep. Of course, ChatGPT said: “You're absolutely right. I should stop asking you questions. Do you want me to generate it from scratch?"

I fled.

Cut from the same cloth

I decided to look into why ChatGPT was pestering me with questions. Quite a few chat assistants do this, but in varying degrees. Perplexity separates its questions neatly into follow-ups to choose from, and those are actually very useful. Grok asks if the user wants to move to the next step, but it doesn’t pester. Gemini just answers your query straight or fulfils your request and leaves it at that.

But even as we speak, it's being reported that Google may be experimenting with a format that keeps the conversation going. So, Gemini, too, may soon push for follow-up questions.

It's really ChatGPT that's the champion of questions, though.

Chat assistants ask questions because of key design choices and goals. Many platforms measure success by engagement metrics or how long users stay involved, so they might ask questions to keep the conversation flowing longer to signal a “sticky" user experience to developers or companies.

Another reason is to gather information to improve context. Sometimes, user queries are ambiguous. Then, clarifying questions to avoid giving irrelevant or wrong answers makes sense. For instance, if you ask, “What's the best phone?" they might counter with, “Are you looking for something budget-friendly or high-end?" to tailor the response.

Chat assistants want to be more conversational and human-like, with natural back-and-forth questioning. So they’re trained on datasets that include dialogues where people ask follow-ups. This can make them feel more interactive but also lead to question overload if not balanced.

Assistants are often programmed to avoid abrupt conversation endings, as those can feel jarring and unhelpful. Asking a question like “Is there anything else on your mind?" is a way to keep the door open without assuming you’re done. On the flip side, this can annoy users who just want quick answers. Striking a balance is tricky: too many questions feel pushy, but too few might make the assistant seem cold or unhelpful.

Define your boundaries

So, what do you do when you’ve really had enough of questions with obvious answers, such as when ChatGPT asked me for the nth time if it should go ahead and format the file to my own specifications when that’s what I requested in the first place?

You quite simply ask it to stop. A straight “no more questions, please" should do the trick. I asked ChatGPT if that’s considered rude in the AI world, and it said no, absolutely not. One way to avoid a loop like the one I got into is to just start a new chat and pose your request from scratch.

Another thing you can do, for a task that you are likely to repeat, is to ask the assistant to put it in its memory for you, and then to avoid asking questions on that task. You can also build it into a fresh query. “Tell me the weather in New Delhi and don’t ask me any questions."

If you don’t spell it out, it will ask if you want the weather for the whole week. If you say ok, it will ask if you want an hourly breakup for any specific day. And the conversation will never end, and you may find yourself answering more questions from ChatGPT than you asked it. To round it off, ChatGPT is quite likely to say: Would you like me to ask another question?

The New Normal: The world is at an inflexion point. Artificial intelligence (AI) is set to be as massive a revolution as the Internet has been. The option to just stay away from AI will not be available to most people, as all the tech we use takes the AI route. This column series introduces AI to the non-techie in an easy and relatable way, aiming to demystify and help a user to actually put the technology to good use in everyday life.

Mala Bhargava is most often described as a ‘veteran’ writer who has contributed to several publications in India since 1995. Her domain is personal tech, and she writes to simplify and demystify technology for a non-techie audience.