CM-04: Articulation, Curiosity, and the Art of Prompting (Prompt Architecture)

Core Theme: AI does not reward cleverness; it rewards clarity. If you cannot question what you need with the unquenchable curiosity of a small child, and then translate it into unambiguous words, even the most powerful supercomputer on Earth cannot give you the result you want.

1. The Hidden Secret of Prompting: The Toddler's Curiosity

People often come to me and ask: "Sir, how do I become a good prompt engineer? What kind of coding do I need to learn for that?"

My answer astonishes them: "To become a good prompter, what you must do is not learn computer code — it is to become a three-year-old child again!"

If there is a three- or four-year-old in your home or in a relative's house, you have lived this. That child spends the entire day asking questions. There is not a single moment without a question:

• "Amma, why is the sun round? Why is the moon round?"
• "Why isn't the sun there at night? Why doesn't the moon come out in the day?"
• "If I climb to the top of Sigiriya and drop my smartphone, will it float away? Will it fall and break? Will it hit the rock on the way down? Will the wind carry it off? If I jump after it, can I catch it mid-air? Will I float too? What will happen to me?"

Figure 5.1: The Childlike Curiosity — the unquenchable curiosity with which a small child looks at the world is the origin of prompt engineering.

Why does this child ask questions without end? One reason only: curiosity.

Unfortunately, as we grow older, our traditional education system and our society blunt that curiosity. We are trained to stop asking and to accept, silently, whatever we are given.

To win in the age of artificial intelligence, you must reawaken the curiosity you buried. Prompting is nothing else: looking at a problem with curiosity, hunting for its root, and asking a continuous, logical chain of questions. The person who questions with curiosity becomes, by nature, an excellent prompter.

Figure 5.2: Art of Inquiry — questioning continuously to find a problem's root causes.

Figure 5.3: Stay Curious — keeping the flame of curiosity from going out.

Figure 5.4: Question Everything — the discipline of questioning everything to identify the true root of problems.

2. Articulation: Jordan Peterson's Philosophy (Jordan Peterson on Articulation)

Curiosity is the fuel. But to steer that curiosity toward its true target, you need the rudder called articulation — the clear expression of ideas.

The view of the Canadian psychology professor Dr. Jordan B. Peterson on articulation should be a foundational compass for all of us in the AI era:

Figure 5.5: Jordan B Peterson on Articulation — how human expressive power is sharpened by choosing words carefully, speaking what you believe to be true, and reading and writing deeply.

"You become more articulate by choosing your words carefully, by practicing saying what you believe to be true, by reading and writing your way into the depths of your own thought, and by treating the word with immense respect. The clearer your expression, the more powerful your thinking; the more powerful your thinking, the more you prevail in the world."

The most powerful person in the AI era is not the one who memorized computer code. It is the person with the articulate power to tell the machine what they need — in exact words, with zero ambiguity.

3. The 5 Essential Human Skills Powering the Prompt

To obtain a flawless professional document from an AI model, there are five fundamental skills that the human — not the machine — must possess (Figure 5.6):

Figure 5.6: The 5 essential human skills of prompt engineering — critical reading, explicit writing, structured logic, critical thinking, and precise articulation.

The Anatomy of Figure 5.6's Five Human Skills:

1. Critical Reading (Context Ingestion): Reading the problem, the complaint, or the draft bill deeply, without laziness. Asking an AI to solve something you have not read is one blind man asking another for directions.
2. Explicit Writing (Instruction Precision): Stripping away unnecessary words and converting the exact goal into direct, clear, imperative instructions.
3. Structured Logic (Sequential Tasks): Not demanding a complex administrative task all at once, but decomposing it into a logical sequence of sub-routines: inspection → eligibility verification → calculation → recommendation.
4. Critical Thinking (Logic & Analysis): Vigilantly examining the AI's answer for logical gaps, conceptual contradictions, or false inferences.
5. Precise Articulation (Prompt Syntax): Arranging the role, context, constraints, and format so cleanly that the machine cannot possibly misunderstand.

4. The Four Pillars of Precise Articulation (Precise Articulation Framework)

A single ambiguous word in a document can cost a citizen their land title or their pension, collapse a business contract, or even lead to a High Court case. Figure 5.7 shows the four core principles to follow when constructing a prompt:

Figure 5.7: The four pillars of precise articulation in prompt engineering — Zero-Ambiguity, Negative Constraints, the Intent-Action-Result (IAR) audit model, and Semantic Precision.

The Deep Anatomy of Figure 5.7's Four Principles:

1. Pillar 1: The "Zero-Ambiguity" Framework: * Remove from the prompt any subjective word that could carry multiple meanings. * Avoid: "Write this very professionally." ("Professional" is a vague word that means different things to different people.) * Adopt: "Write this in a direct, authoritative tone, leading with action-oriented verbs, in simple sentences a person with an ordinary school education can understand."
2. Pillar 2: Constraints as Scaffolding: * Telling an AI what it must not do is as decisive as telling it what to do. * Negative constraints: "Do not use passive voice. Do not use empty corporate buzzwords like 'synergy' or 'optimization.' No more than two sentences per paragraph." * Structural constraints: "Format the output as a 3-column table (Cost, Risk, Velocity)."
3. Pillar 3: The "Intent-Action-Result" (IAR) Audit Model: * Before you press Enter, audit your own prompt against the three IAR criteria:
   • Intent (Goal): What is the institutional purpose of this task? (E.g., "Persuade the Ministry of Finance to approve the budget allocation.")
   • Action (Task): What is the specific mechanical task delegated to the model? (E.g., "Compare the expenditure reports of the 3 attached Divisional Secretariat offices.")
   • Result (Outcome): What form must the final winning document take? (E.g., "A one-page executive summary with a clear 'approve / do not approve' recommendation.")
4. Pillar 4: Semantic Precision (Anchoring Latent Space): * Using the field's exact technical or legal terminology — instead of everyday language — anchors the model's neural attention in the relevant expert domain. * Vague: "Write good software code" → Precise: "Write a stateless functional component in TypeScript, using memoization for efficiency." * Vague: "Make a good marketing plan for farmers" → Precise: "Using the Jobs-to-be-Done (JTBD) framework, identify the core struggle of the rural farmer applying for the fertilizer subsidy, and design a communication strategy that speaks to it."

5. Prompt Architecture: The 5 Components of a Complete Prompt (Universal Prompt Architecture)

A prompt built to analyze complex professional documents — case files, gazettes, lesson plans, business reports — is not a sentence or two. It is a deliberately designed piece of architecture:

Figure 5.8: Prompt architecture (Full Architecture) — Role, Context, Task, Constraints, and Output Format.

The Deep Anatomy of Figure 5.8's Five Layers:

1. Layer 1: Role and Persona (<role>): The expert identity assigned to the model. E.g.: "Act as an Additional Secretary with 20 years in the Sri Lanka Administrative Service, expert in the Establishments Code and government financial regulations."
2. Layer 2: Context and Background (<context>): The background information, relevant circular numbers, and parties involved. E.g.: "A dispute has arisen between the local council and the traders' association over the removal of unauthorized shops in the Divisional Secretariat area. Under Circular No. 12 of 2024..."
3. Layer 3: The Specific Task (<task>): The direct operation to perform. E.g.: "Draft a 5-step mediation action plan to bring the two parties to a settlement."
4. Layer 4: Constraints and Rules (<constraints>): What must not be done, legal limits, and word count. E.g.: "Propose nothing that exceeds the powers of the Local Government Act. Do not include inferred laws. Limit to 300 words."
5. Layer 5: Output Format (<output_format>): The shape the answer must take. E.g.: "Present with headings, numbered steps, and a closing Markdown table of [Risk | Responsible Officer | Remedy]."

6. Idasara Academy and the Skill of Articulation (Cross-linking to Learn to Learn)

This expressive power is exactly what we emphasize, relentlessly, at Idasara Academy (Meta-Learning Superpower).

A student fails an exam — or a graduate fails a job interview — not because they lack the facts, but because they cannot articulate what they know clearly and in an organized way (Certificates vs Capability).

Whether you are a student, a teacher, a parent, or a professional: practicing this skill of articulation in your daily writing and in commanding AI tools will make you a professional without equal in your field.

In the next module, building on these principles, we study the 9 advanced prompting methods — and a hands-on case study of how they were put to work inside a real institution: the "Promptathon".