In partnership with
Give ChatGPT campaign results, customer feedback, or a rough strategic opinion and it can turn it into polished analysis in seconds.
But polished doesn't mean proven.
AI can easily blur the line between what the evidence actually shows, what you think it means, and what you recommend doing next. A reasonable assumption can come back sounding like an established fact.
Before you put AI-generated analysis into a report, strategy or presentation, get ChatGPT to audit the claims first.
Try this prompt:
You are a critical thinking and evidence analyst. Review the draft below and audit every material claim before I use it.
Create a table with:
- Claim
- Classification: fact, inference, assumption or recommendation
- Evidence supplied
- What would be needed to verify it
- Confidence: high, medium or low
- Safe to present as written: yes or no
Be strict. Do not treat plausible explanations as facts. If the evidence only shows correlation, do not imply causation. Flag any claim that sounds more certain than the evidence supports.
Then rewrite the draft so that:
1. Facts are stated clearly
2. Inferences and assumptions are presented with appropriate uncertainty
3. Recommendations are clearly separated from findings
4. Unsupported claims are removed or qualified
Draft:
[PASTE REPORT, ANALYSIS OR RECOMMENDATION]
The goal isn't to make your work sound less confident. It's to make sure your confidence is earned.
Happy prompting!
Your Competitors Just Hired an Entire GTM Team. No Humans.
AI agents are doing the work of full GTM teams. Pipeline, content, customer service. The startups using them aren't waiting for headcount approvals.
Get the free Practical Guide to Agentic GTM for Startups and see exactly where to start.