Lauren Fox5 ChatGPT Power Words
Five ways to sharpen a ChatGPT prompt, with copyable examples for LOSSAVERSION, SHADOWMOTIVE, CONTRASTMODE, INCENTIVEMAP and SPECIFICITY.
These five words are shorthand for useful ways to frame a question. Give ChatGPT the instruction behind the word as well as your situation, so it knows exactly what you want. Copy an example below and replace the details in square brackets.
How to use the words
Open a ChatGPT conversation, choose the word that fits your question and paste its full prompt. Include the goal, the options and any constraints that matter. These are prompt labels used in this guide, not special settings you need to unlock.
1. LOSSAVERSION
See the cost of doing nothing.
LOSSAVERSION: I am deciding whether to [decision]. My current situation is [context]. Show what I could lose by doing nothing over [time period], including time, opportunities and effort. Compare this with the costs and risks of making the change. State assumptions and do not invent figures.2. SHADOWMOTIVE
Explore what may sit underneath your question.
SHADOWMOTIVE: My question is [question]. Here is the context: [context]. Suggest possible needs, concerns or goals behind this question. Treat them as hypotheses, not facts about me. Explain what supports each possibility and ask me questions to check which fits.3. CONTRASTMODE
Put two options side by side.
CONTRASTMODE: Compare [option A] with [option B] for someone whose goal is [goal] and constraints are [constraints]. Use a side-by-side table covering the differences that matter to my decision. Explain when each option makes more sense and what missing information could change your conclusion.4. INCENTIVEMAP
Understand who benefits from what.
INCENTIVEMAP: Help me understand this situation: [situation]. List the people or organisations involved, what each may gain or lose, and how those incentives could shape their advice or behaviour. Separate known facts from possible explanations. Do not assume a hidden motive without evidence.5. SPECIFICITY
Turn vague advice into usable detail.
SPECIFICITY: Make your answer to [question] concrete. My context is [context]. Give specific steps and a worked example. Use numbers where there is a sound basis for them; label illustrative numbers as hypothetical. If you need information from me to be precise, ask instead of making it up.Try two together
For a decision, CONTRASTMODE and SPECIFICITY work well together. Define both instructions in the prompt rather than relying on the labels alone.
CONTRASTMODE + SPECIFICITY: I am choosing between [option A] and [option B]. My goal is [goal], and my constraints are [constraints]. Compare them side by side, then give a worked example showing how each would look in practice. Label assumptions and hypothetical numbers. End with the one question I should answer before deciding.Get a better second answer
If the answer is still vague, add a real example from your situation and say which part needs more detail. If it guesses at your motives or someone else's, ask what evidence supports the claim. The useful part is the instruction and context you give it, so keep those even when you shorten the label.
More guides are on the way
Follow Lauren for new AI tools, practical prompts and the next guide.