daBongo LMS AI Training Courses

Lesson Content

The "before you answer" technique.

For any complex or important prompt, use a two-step approach:

Step 1: "Before you answer this question, summarize what you are being asked and what assumptions you are making about my situation."

Review the summary. If Meta AI misunderstood, correct it before it generates the full response.

Step 2: Once the understanding is confirmed, say: "That's right – now provide the full response."

This technique prevents investing in a long response that misses the mark.

Assumption surfacing after a response.

When a response feels generic or slightly off: "What assumptions did you make about my situation in this response that I should know about? What context would change your answer significantly?"

This surfaces the implicit decisions Meta AI made – often revealing why the response does not quite fit.

Progressive conversation framing.

For complex tasks that benefit from building up context:

1. First message: "I am going to ask you to help me with [complex task]. Before I give you the details, tell me what information would be most helpful for you to have."
2. Second message: Provide the requested context.
3. Third message: The actual task request.

This produces better outcomes for complex, high-stakes tasks where getting it right on the first try matters.

The calibration check.

After a long conversation or when returning to a complex topic: "Briefly confirm: what do you understand my goal to be, and what have we established so far about my situation?"

This catches context drift – cases where Meta AI's understanding has shifted during the conversation.