Most people use ChatGPT to polish an offer. Ask it to review your proposal and it tells you the tone reads well and suggests tightening a sentence. That is not the useful part. What you need before you send a number is the argument coming back at you.
The default setting is agreeable
Left alone, ChatGPT gives feedback. It softens. It lists concerns politely and stops short of the sentence that would actually land in a negotiation. You have to name the posture or you get a review instead of an opponent.
The fix is one line at the end of the prompt: be adversarial, not diplomatic.
Ask for three specific things
Three arguments against. Their strongest case, not a polite list of concerns.
One counter-number. Where this actually lands once they push back.
One weak point. The line in your reasoning they attack first.
The counts matter. Ask for "objections" and you get a vague paragraph. Ask for three arguments, one number and one weak point and you get something you can prepare against.
The prompt
Here is my offer: [paste it]. Take the other side. Give me the three strongest arguments against accepting this, the number you would counter with, and the one weak point in my reasoning you would attack first. Be adversarial, not diplomatic.Open the web version to copy this cleanly. Copying out of an inbox tends to drag stray line breaks with it.
What to do with what comes back
The counter-number is a guess, not intelligence. ChatGPT has no idea what the other side's budget is, so treat it as a stress test of your own reasoning rather than a forecast.
The weak point is the part worth your time. If ChatGPT finds the soft line in your logic in four seconds, the person reading your email will find it too.
This works on salary conversations, vendor quotes, scope pushback, and any proposal where you will have to defend a number out loud.