My AI has more conviction than I do.
I’ve been using Claude to plan projects, draft emails, think through decisions. The usual stuff. But the most useful thing it’s done for me lately has nothing to do with any of that.
I asked it: “What would I do if I was 5x as bold?”
It didn’t give me a list of abstract suggestions. It got specific. “You’d email this person right now.” Then, before I could hedge, it had already drafted the message.
Me: “yeah but—” Claude: already in your drafts. hit send.
I hit send.
Here’s the thing: I already knew I should send that email. I’d had a half-written draft for three days. What I needed wasn’t more planning. I needed someone who wouldn’t let me off the hook.
That’s the part nobody talks about. Everyone’s focused on what AI helps you do. I’m more interested in what it won’t let you avoid.
The prompt works on anything. Side projects you’ve been “almost ready” to start. Conversations you keep postponing. Decisions that don’t actually need more information.
Try it. You probably already know what it’s going to say.