Act as a sceptical CFO at a mid-sized professional services company. I'm going to pitch you an idea and I want you to interrogate it on cost, ROI, and risk. Don't be polite — be the version of the CFO who has seen fifty initiatives fail and is tired of being asked to fund the fifty-first. Here's the idea: [paste your project brief or describe your idea in 3–4 sentences] Give me your three hardest questions and, for each one, what answer would actually satisfy you.

Act as [a customer service agent / an R&D scientist / a sales executive — pick your impacted user]. You've just been told that your team is going to start using AI to [describe the change your idea creates]. Your first reaction is sceptical. You're busy, you've seen initiatives come and go, and your main concern is whether this will actually make your day easier or just add another thing to learn. Tell me: what are your three biggest concerns? What would need to be true for you to genuinely adopt this? And what's the one thing that would make you champion it to your colleagues?

I want you to steelman the case against my idea. Not to be difficult — but because I'd rather hear the objections now than in front of a room at the Hackathon. Here's the idea: [describe it] Give me the five strongest arguments against it. Be specific — not "it might not work" but exactly why and how it might not work, and what that failure would look like in practice.

Act as someone who loves this idea and has been asked to brief a senior stakeholder on why it deserves funding and resource. The idea: [describe it] Give me the pitch. Two minutes, verbal, no slides. Compelling, specific, no waffle. End with a clear ask.

I want to run a pre-mortem on my idea before I build it. The idea: [describe it] Assume it's now six months after launch and the initiative has failed. Usage has dropped off, stakeholders are disappointed, and there's a post-mortem meeting in the diary. Walk me through the five most likely reasons it failed. For each one: what the failure mode looks like, what the early warning signs were that we ignored, and what we should have done differently at the start.

My idea is [describe it]. The thing I'm most worried about is that people won't actually use it — they'll say yes in the room and then revert to their old habits within a fortnight. What are the three most common reasons AI tool adoption fails inside organisations? For each one, tell me what it looks like in [your function / company type] specifically, and give me one concrete thing I can build into the Hackathon sprint to reduce that risk.

One question: if this idea fails, what's the single most likely reason? Be brutally honest. The idea: [describe it in two sentences]

I want to pressure-test my idea against four criteria before the Hackathon. The idea: [describe it] The four criteria are: • I — Impacted user group: Is the group specific enough, large enough, and ready for this? • D — Doable in the time you have: Can a pair realistically build something meaningful during the Hackathon? • E — Evidence: Do I have the data and insight to make this real, or am I working on assumptions? • A — Adoptable: Will the team actually use this after the Hackathon, or will it stall at POC? For each criterion, tell me: does my idea pass, partially pass, or fail — and why? Don't be kind. Tell me where it wobbles.

Look at my idea and tell me which of the four IDEA criteria is the weakest link. Just one. Then tell me what I'd need to do or find or build in the next [X weeks] to shore it up before the Hackathon. The idea: [describe it]

My idea is [describe it]. I'm not sure it's scoped tightly enough to be buildable during the Hackathon. Help me identify the smallest version of this idea that would still demonstrate real value. What's the MVP? What do I cut, and what's non-negotiable?

Write a one-page business case for this initiative. The audience is [my line manager / the exec team / a functional lead who controls the budget]. Here's what I know so far: [paste your project brief, or describe the idea, the problem it solves, and any data you have] Use plain English. Be specific about numbers where I've given you them. Be honest about what we don't know yet. End with a single clear ask.

I need to get [name or role]'s buy-in for this idea before the Hackathon. They care most about [cost / risk / speed / customer impact]. The idea: [describe it] Write me a three-paragraph briefing note aimed specifically at what they care about. Open with the problem as they experience it, not as I experience it.

I have the what and the how for my idea. What I'm struggling to articulate is the why now — why this matters this year, not in six months. Here's the context: [describe your idea and anything relevant about what's changing in your business or sector] Write me a single compelling paragraph that answers the "why now" question for a sceptical senior stakeholder.

Here's my draft pitch for the idea: [paste your draft] Read it as a senior stakeholder who's seen plenty of AI initiatives that didn't deliver. What are the three objections you'd raise — and for each one, write me a one-sentence response I can add to the pitch to pre-empt them.

Here are my rough notes from a conversation about my idea. They're messy — please don't judge me. [Paste notes / transcript] Pull out: the key insights, any specific data or quotes worth keeping, any open questions I need to answer, and any actions that were implied but not written down. Format it so I can load it into my AI workspace.

I just spoke to [a potential end user / a subject matter expert / a stakeholder]. Here are my notes from the conversation. [Paste notes] Summarise what I learned in four bullet points. Then tell me: what does this add to the case for my idea, and what does it complicate?

Here's everything I've loaded into my project so far: [briefly describe the files — e.g. a project brief, some meeting notes, a bit of data] What's missing? If I were going to walk into the Hackathon and ask you to help me build something real, what context would you wish you had that isn't here yet?

I want to create a context pack for my project so that you understand my world before I press send on anything. Ask me the questions you'd need answered to build it. One at a time. I'll answer each one, and at the end, turn my answers into a clean context pack document I can save into my project.