Make AI argue against a plan I've already decided on

I've decided to do this: [YOUR PLAN, IN TWO OR THREE SENTENCES — what you're doing and why]

Here's what I know: [THE FACTS YOU'RE WORKING FROM — budget, timing, who's involved, what you've already committed]

Your job is not to help me do it. Your job is to try to talk me out of it, properly.

1. THE STRONGEST CASE AGAINST. Not a list of risks — an argument. Make it the version a smart person who likes me would make if they were genuinely worried.

2. THE ASSUMPTION HOLDING IT UP. Every plan rests on one thing being true. Name the load-bearing one — the belief that, if wrong, makes the rest collapse rather than just wobble.

3. HOW I'D KNOW, CHEAPLY AND SOON. What's the smallest, fastest test that would tell me that assumption is wrong? It should cost days, not months.

4. WHAT I'M OPTIMISING FOR THAT I HAVEN'T ADMITTED. Sometimes the real driver is looking decisive, or avoiding a conversation, or sunk cost. If you can see one, say it.

5. THE VERSION THAT SURVIVES. Given all of the above — what's the smaller, more reversible plan that gets most of the upside?

Then stop. Do not close by reassuring me the plan is basically sound. If you genuinely can't build a case against it, say "I can't argue against this, and here's why" — but try properly first.

One more thing: use anything you already know about me from our past chats and anything you're connected to. If you can find it yourself, don't make me look it up.