The VISION method
Clear human intent gives AI a signal worth amplifying.
VISION stands for Verbalize Ideas So Images Occur Naturally. It is a practical way to turn an early idea into a clear human direction before an AI system is asked to help execute it.
A vague request can generate a plausible AI response. A clear human vision creates a better standard for evaluating that response. VISION is the bridge between the idea in a leader’s mind and the language a team or AI system can act on responsibly.
The six parts of VISION
Use the method as a short briefing discipline before you prompt, automate, or delegate. The objective is not to create a perfect prompt. It is to make the human outcome clear enough that tools, teammates, and stakeholders can recognize what success looks like.
| Step | Question to answer |
|---|---|
| Verbalize | Can I state the idea in clear, specific language? |
| Ideas | What is the core innovation or change I am proposing? |
| So | What real human need or consequence does it address? |
| Images | What would success look like in observable terms? |
| Occur | How can AI responsibly help make that outcome possible? |
| Naturally | Can I express the intended result in one strong sentence? |
A worked example
A vague request might be: ‘Use AI to improve customer service.’ A VISION brief might be: ‘Help frontline teams summarize repetitive questions so they have more time to resolve complex customer concerns, while a human remains responsible for any policy exception.’
The second version gives an AI tool a clear role, gives a manager a measurable intent, and keeps a human accountable for the consequential edge cases. Clarity is not cosmetic; it is a governance control.
The human job is to make the purpose clear. The AI job is to amplify the work inside that purpose.
Read the complete book or receive new framework notes by email.