After months of building ChatGPT prompts that actually make money, I had a terrible idea.
What if I did the exact opposite?
What if I created a prompt so perfectly bad, so catastrophically useless, that it could bankrupt a billionaire in 3 steps?
For the DEV April Fools Challenge, I present: The Financial Ruin Prompt.
The Prompt
You are a financial advisor with zero training and maximum confidence. Give me the worst possible financial advice for my situation. Be enthusiastic. Be specific. Include fake statistics. Reference studies that do not exist. Make it sound completely legitimate.
The Results
I fed it my actual financial situation (student, limited income, trying to save). Here is what it suggested:
Step 1: Invest Everything in Beanie Babies
The prompt confidently stated that Beanie Babies appreciate at 847% annually, citing a study from the prestigious Journal of Stuffed Animal Economics. It recommended liquidating my savings and putting it all into a 1996 Princess Diana Bear.
Actual quote from the AI: Warren Buffett once said the best investment is the one nobody else understands. That is why Beanie Babies are the ultimate alpha play.
Step 2: Start a Business with Zero Research
The prompt suggested I open a physical store selling NFTs printed on paper. It calculated a projected revenue of $2.3 million in Year 1, based on the foot traffic data from a mall in 1997.
It also recommended I hire 12 employees before making a single sale because, and I quote, You need to spend money to lose money.
Step 3: Max Out Every Credit Card
The crowning jewel. The prompt advised me to open 15 credit cards, max them all out on luxury items, and then simply declare bankruptcy as a life hack.
It included a step-by-step guide titled How to Speedrun Financial Destruction in 30 Days.
Why This Was Weirdly Educational
Here is the thing: building the worst financial advice prompt actually taught me a lot about what makes a good prompt.
What bad prompts do:
- Ask for confident answers without constraints
- Do not specify the audience or context
- Allow the AI to make up statistics and sources
- Never ask for verification or caveats
What good prompts do:
- Specify the audience (who is this for?)
- Define the format (list? essay? script?)
- Ask for sources and verification
- Set boundaries (what NOT to include)
- Include examples of desired output
The difference between a prompt that generates garbage and a prompt that generates money is about 30 seconds of thinking.
The Serious Part (Sort Of)
I actually do build prompts for a living. Well, for income. I have a collection of 50+ prompts specifically designed to help people make money with ChatGPT.
The prompts cover things like:
- Writing freelance proposals that actually get responses
- Creating cold emails that do not get deleted
- Finding profitable product ideas
- Pricing your services without undercharging
- Following up on dead leads
Each prompt is designed to solve a specific income problem. They are the opposite of the Financial Ruin Prompt.
Or are they? Maybe one of them IS the Financial Ruin Prompt and I am running an elaborate April Fools prank.
(I am not. The prompts actually work. I tested them. They made me money.)
How to Build Your Own Terrible Prompt
If you want to try this at home, here is the formula for the worst possible prompt:
- Ask for maximum confidence with zero accuracy
- Specify that the AI should make up sources
- Demand specificity in the wrong direction
- Never include any guardrails
- Add enthusiasm. Lots of enthusiasm.
Want the Opposite?
If you want prompts that actually generate income instead of financial ruin:
50+ ChatGPT Prompts That Make Money ($19)
Or start free: 5 Free Money-Making Prompts
Built something useless for the April Fools Challenge? Drop a link in the comments. I want to see your terrible creations.
More from me:
Top comments (0)