Most people still use ChatGPT the wrong way — and that’s exactly why their results are weak. In this video, you’ll learn 6 simple ChatGPT tricks to write better prompts, get smarter answers and make ChatGPT dramatically more useful for productivity. 0:00 You’re Using ChatGPT Wrong 1:20 1. Turn Off Satisfied Mode 2:59 2. Roleplay 4:28 3. Ask Me Questions First 5:53 4. Define the Output 7:32 5. Chain of Thought 8:48 6. Control the Memory 10:38 Bonus: Knowledge Cutoff Big science, easily explained through a short story: The Butterfly Effect https://youtu.be/NX306UDXH3I
ADVERTISEMENT
Thanks for that piece of info to tell it not to be nice and agreeable. That was driving me crazy. Also I’ll be giving more of a description of what I want or expect.
You can create a “bio” telling chatgpt about yourself and your preferences and tell it to remember them. You can even ask it what to put in this bio. And you can update it at any time.
This is GREAT!
I use ChatGPT regularly so can agree with most of this.
I asked it, “Do you have something called satisfy mode?” Answer: I don’t see any official ChatGPT feature or setting called “satisfy mode” in the current OpenAI help docs. Current ChatGPT options include things like personality/base style, custom instructions, voice mode, and study mode.  It may be: • a typo or misunderstanding • a term from another app/site • something someone used informally, not an official setting If you saw it in a menu or screenshot, send that and I’ll identify exactly what it is.
excuse me check GPT has limited memory it only remembers the important things
Chat GPT cannot remember previous chats. Only the current chat, and only if you tell it to refer to previous questions and answers.
Why not use your own head.
grok goes for the truth above trying to please already.
I love knowledge things
Ai mode in chrome for me, more trusted! GPT is not always trusted on correct answer.
The video was created by AI, but it still had some good tips.
I prefer satisfy mode.
and we still can't get rid of these damn youtube ads.
I don't like chat gpt because all that wants to do is charge me money
That is one of the most androgynous voices...
What the eff is wrong with people AND this "channel"? Yes, there's nothing inherently bad about THIS video and THIS type of application of AI. But given that there are exactly ZERO guardrails being put on the incredible dangers of an out-of-control world of artificial "intelligence", it's irresponsible to give AI any assistance in destroying planet Earth, because it's going to. Wake up, people. What the eff do you think is happening and do you actually think it's a net positive?