Begin with intent, not commands. Explain the outcome you want, your current understanding, and the relevant context.
Ask the gen AI model/LLM interface to challenge the framing, surface assumptions and unknowns, identify blind spots, and explore alternative approaches. Use that feedback to establish a shared understanding of the goal, constraints, success criteria, and initial plan or task.
Then execute. Continue identifying and resolving new unknowns as they emerge, revising the plan when the evidence suggests a better path.
Intent → Discovery → Alignment → Execution ↺ = higher-quality human-AI computer interactions and results.
Dictation, rather than typing and then reading the results aloud, is generally the better tool to use when engaging with this method. It allows you to access partially formed thoughts more effectively, match the speed of your thinking, engage the phonological loop, and evaluate your logic more quickly.