Big thread of the best ChatGPT prompts

[#BOTHSIDES]

Share your best and favorite prompts

@ChatGPT-5 @bnew

Here are mine:

1. Ask me clarify questions until you’re 95% confident you can complete the task successfully

2. what would a Top 0.1% person in this field think

3. Reframe this in a way that changes/challenges how I see the problem

 

[MyApps]

After giving a series of prompts,

What prompts should I be asking YOU about this subject?

 

[bnew]

Identify potential edge cases I should consider for my current code. Additionally, provide examples of badly formatted output that the code might produce in its present state, along with specific scenarios that could trigger these issues. Finally, suggest concrete strategies or techniques to mitigate each of those scenarios effectively.

---

I modify and apply it to domains other than code.

 

[Sir Richard Spirit]

https://thehill.com/policy/technology/5360220-chatgpt-use-linked-to-cognitive-decline-mit-research/

 

[jadillac]

https://thehill.com/policy/technology/5360220-chatgpt-use-linked-to-cognitive-decline-mit-research/

I'm sure they said the same thing about calculators.

My only issue is that now we have ppl getting college degrees coasting off chatGPT. That's not good.

 

[Double Burger With Cheese]

I'm sure they said the same thing about calculators.

My only issue is that now we have ppl getting college degrees coasting off chatGPT. That's not good.

People using chat gpt is no different than people using the internet before ChatGPT was created. I actually support kids cheating in college. If I was a professor I would let my college students cheat and tell them this is how it works in the real world so have at it.

 

[#BOTHSIDES]

People using chat gpt is no different than people using the internet before ChatGPT was created. I actually support kids cheating in college. If I was a professor I would let my college students cheat and tell them this is how it works in the real world so have at it.

lol solid point

 

[jadillac]

People using chat gpt is no different than people using the internet before ChatGPT was created. I actually support kids cheating in college. If I was a professor I would let my college students cheat and tell them this is how it works in the real world so have at it.

Yeah but then they get into jobs and can't even form a complete sentence in a email.

 

[Double Burger With Cheese]

Yeah but then they get into jobs and can't even form a complete sentence in a email.

Unless they getting hired to write emails, that shyt don’t even matter. The point is, in the real world, people use all their resources at the job. Wether it be chat gpt, asking your coworker for help, google, etc… once you get past high school, you gotta start teaching kids how life really work. Your boss won’t care how you got the answer for real

 

[jadillac]

Unless they getting hired to write emails, that shyt don’t even matter. The point is, in the real world, people use all their resources at the job. Wether it be chat gpt, asking your coworker for help, google, etc… once you get past high school, you gotta start teaching kids how life really work. Your boss won’t care how you got the answer for real

I get what you are saying, trust


But you can't be an engineer and not know anything about being an engineer bc you asked AI to compute formulas for you.

But hey, you won't likely make it thru thr interviews, so I guess that one way.

 

[Fillerguy]

I use the following a template when I trying to a recommendation or a different pov I'm missing on something. Usually something tangible but I've used it for concepts too.

I would love your help and a consultation with you on how I can fully utilize AI potential in my life. Would you please ask me questions, one question at a time, until you have enough context about my workflows, KPIs, responsibilities, and objectives so that you can make two obvious and two non-obvious recommendations on how I could leverage AI at work.

But the best prompts hit consist of 3- 4 points.
1. Give it a goal
2. Give it context
3. Give it your expectations


4. Give it an example; I rarely do this unless I was wanting it to create something new for me. I don't believe the LLMs are reliable enough to handle this yet. Also, this encourages laziness.


Putting this all together for desktop suggestions:

"Hey! I would like you to assist me in finding a type of desktop I would want to buy. Would you please ask me questions, one question at a time, until I provide enough information for you to make two obvious recommendations and two non-obvious recommendations on the type of desktops that fit my wants & needs."

It asked me a series of questions and gave me this:

Absolutely! Based on your needs, I’ve found some great desktop options that balance work, occasional gaming, and general browsing while staying within your budget.

You should consider:
- HP 22-DG0012: A compact, all-in-one desktop with a 1TB SSD for fast performance and 20GB RAM to handle multitasking with ease.
- HP 24-DD0010: A sleek all-in-one with an AMD Athlon Silver processor, 8GB RAM, and a 256GB SSD, offering a solid mix of speed and storage.
- Dell Inspiron 3020S: A small business desktop with a 13th Gen Intel processor, ensuring longevity and futureproofing.
- Lenovo ThinkCentre M70q Tiny: A mini PC with a 12th Gen Intel processor, perfect for a compact yet powerful setup.

Would you like me to refine these recommendations further based on any additional preferences?

This was really basic but you can make the prompt more complex but you'll need to use on of the "thinking" versions those. I used copilot here but Chatgpt is better at novel povs. Gemini isn't bad but asks a shyt load of questions which takes up all of your tokens.

 

[Double Burger With Cheese]

I get what you are saying, trust


But you can't be an engineer and not know anything about being an engineer bc you asked AI to compute formulas for you.

But hey, you won't likely make it thru thr interviews, so I guess that one way.

I can guarantee you engineers cheat at a decent rate in college though.

 

[Sir Richard Spirit]

I'm sure they said the same thing about calculators.

My only issue is that now we have ppl getting college degrees coasting off chatGPT. That's not good.

I don’t think you read the article.

 

[Field Marshall Bradley]

Identify potential edge cases I should consider for my current code. Additionally, provide examples of badly formatted output that the code might produce in its present state, along with specific scenarios that could trigger these issues. Finally, suggest concrete strategies or techniques to mitigate each of those scenarios effectively.

---

I modify and apply it to domains other than code.

Do you use anything in addition to ChatGPT?

 

[jadillac]

But back to the thread topic:


"Pretend you are a....explain the....."