Most people use ChatGPT like a smarter search engine: type a question, get an answer, move on. But after months of using AI tools every single day, I’ve realized something that a lot of people miss — tiny changes to the way you phrase prompts can completely transform the quality of the responses you get back.
Eventually, you start to figure out these things yourself through trial and error, but it's a lot easier if somebody gives you a head start. So, here are the best hacks I've discovered that produce better, more meaningful results from ChatGPT, Gemini, Claude, and other AI chatbots.
The best part is that they’re not long, complicated prompts you need to copy and paste constantly. Instead, they’re simple phrases you can add either at the start of a chat or at the end of your request — almost as an afterthought — but they can make a surprisingly big difference to the kind of answers you get back.
1. Make no assumptions
One thing you quickly notice when using ChatGPT is that it can sometimes go off in strange directions because it only has the information you’ve written in your prompt. To fill in the gaps, it makes assumptions — and those assumptions aren’t always correct.
Quite often, the AI will ask clarifying questions before answering, but you can make sure it happens more consistently by adding these two sentences to the end of your prompt.
Just type: “Make no assumptions. Ask me for clarification before you begin.”
It’s a tiny change, but it can produce far more accurate and on-point responses.
2. Make me a prompt to have you...
You can actually ask AI to create prompts for you to use on itself. As strange as it sounds, AI is actually very good at figuring out the best wording to get stronger results from itself.
For example, using ChatGPT, I tried: “Make me a prompt to have you tell me the most important news stories with an emphasis on sport.”
ChatGPT then produced an impressively detailed prompt for me to paste back into a new chat — one that not only gathered the day’s major sports stories, but also highlighted online arguments, controversies, and underreported talking points.
Here it is:
Act as my personal news editor. Give me a concise roundup of the most important news stories from the last 24 hours, ranked by significance and public impact. Prioritize sport heavily — especially major football, combat sports, Formula 1, tennis, and big international events — but also include any genuinely major world news, tech, politics, science, or culture stories that people will likely be talking about today.
For each story:
• Give a punchy headline
• Explain why it matters in 2–4 sentences
• Include key context or consequences
• Flag if the story is controversial, surprising, historic, or likely to dominate social media
• Separate confirmed facts from speculation or rumors
Structure it like a smart morning briefing rather than a dry newswire.
Finish with:
• “What everyone online is arguing about”
• “One underreported story worth watching”
• “Biggest sports talking point of the day”
Keep the tone sharp, informed, and readable — like a mix of a newsroom briefing and a really good sports desk editor.
3. No em dashes
Typing “No em dashes” is a surprisingly effective instruction if you’re tired of ChatGPT constantly using em dashes (—) in its writing.
There’s nothing inherently wrong with em dashes — I use them myself — but ChatGPT is extremely fond of them, and once you notice it, you start seeing them everywhere.
If you never want to see another em dash again, try adding something like: “Remember to never use em dashes at all.”
If memory is enabled, ChatGPT may even remember the preference long-term.
4. Act like a brutally honest mentor
If you want unfiltered feedback on a business idea, article draft, creative project, or long-term plan, try starting your prompt with: “Act like a brutally honest mentor.”
This tends to push ChatGPT away from its usual overly supportive tone and toward more critical, direct feedback. The results can genuinely be useful because the AI starts pointing out weaknesses, blind spots, and unrealistic assumptions instead of simply encouraging everything.
That said, be warned: it can occasionally feel a little brutal.
5. Explain this to me like I’m five
There are countless versions of this prompt floating around online, but the variation I always come back to is simply: “Explain this to me like I’m five.”
For complicated subjects, it’s one of the fastest ways to cut through jargon and understand the core idea underneath. Whether it’s quantum physics, mortgages, AI models, or tax systems, forcing ChatGPT to simplify concepts often reveals whether something actually makes sense or was just buried under complicated language.
Some subjects are just easier to understand when they’re explained like a bedtime story.
Getting better results
The funny thing about AI chatbots is that most people only use a tiny fraction of what they’re capable of. A lot of the difference between a frustrating answer and a genuinely useful one comes down to how you phrase the request.
That’s why these little prompt hacks matter. None of them are complicated, but they change the way ChatGPT responds — whether that means asking better questions, giving more honest feedback, simplifying difficult ideas, or tailoring its writing style to what you actually want.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.