Skip to content
sinc-LLMBook a 30-minute call

Prompt Specification and AI Hallucination Review

An essay using a signal-and-noise analogy to discuss underspecified prompts and the author's proposed prompt structure.

On this page
By Mario Alexandre March 23, 2026 12 min read Beginner Signal QualityAI Myths

Every day, millions of people type something vague into ChatGPT, Claude, or Gemini. They get a confident answer full of made-up facts. Then they screenshot it, post it on Twitter, and write: "AI is hallucinating again."

I have watched this happen for 3 years. It is wrong every single time.

The model did not hallucinate. The model did exactly what it was built to do. It took your signal, ran it through billions of settings, and gave you the most likely answer for what you typed. The problem is that what you typed was almost nothing. You whispered a 10-word sentence into a machine with 175 billion parameters and expected it to read your mind.

That is not hallucination. That is signal failure.

What Hallucination Actually Is

In signal processing, there is a word for what happens when you sample a signal too rarely: aliasing. When you take too few samples, the rebuilt signal has fake frequencies that were never in the original. The output looks real. It sounds right. But it is a side effect of bad input, not a broken algorithm.

The same thing happens with AI prompts. Your request is a rich signal with at least 6 separate information bands: who should answer (PERSONA), what situation you are in (CONTEXT), what facts matter (DATA), what rules apply (CONSTRAINTS), what the answer should look like (FORMAT), and what you actually want done (TASK).

When you type "Write me a marketing strategy," you have given the model 1 of those 6 bands: TASK. The other 5 are empty. The model must fill them in from patterns it learned during training. That filling-in is what people call hallucination. In signal processing, we call it aliasing. The math is the same.

x(t) = Σ x(nT) · sinc((t - nT) / T)

The sinc reconstruction formula motivates this framework’s signal analogy. PERSONA, CONTEXT, DATA, CONSTRAINTS, FORMAT and TASK provide a checklist for prompt details. The analogy does not turn missing fields into a measured fabrication rate or prove that complete fields produce correct answers.

The GPS Analogy

You would not type "restaurant" into Google Maps and then blame the app when it takes you to a random Applebee's in another state. You would know you gave it too little to work with. The app needs more: what kind of food, what city, what price, what time, whether you need parking.

An AI model needs the same kind of detail. The difference is that Google Maps will stop and ask for more. An AI model will not. It was built to always give an answer, no matter how little you typed. It was built for mass use. It works whether your input is great or nearly empty.

A fluent answer can conceal assumptions that your prompt never specified. Check those assumptions against your intended task and the supplied evidence. Prompt gaps are worth investigating, but they do not account for every model error or establish how much of an answer is guesswork.

A prompt-format comparison needs the same workload, model settings, supplied sources, and output checks. Record factual errors separately from the prompt’s field coverage. Related: paper.

Review prompt completeness and output correctness separately. The number of specified fields alone does not establish a hallucination result.

Why Your Prompt Is Whispering

Most prompts fail for one reason. People do not understand what AI actually is. They treat an AI like a coworker. They type the way they would talk to a friend: casually, with shared context and unspoken assumptions. That works with people because people share your history, your workplace, your project, your culture.

An AI model shares none of that. It has only what it learned during training. Every hidden assumption in your prompt is a gap the model fills from that pool. Every "you know what I mean" is a gamble. Every rule you leave out is a fence the model will walk past because it does not know the fence is there.

You are whispering because you think AI reads between the lines. It does not. It reads tokens and calculates probabilities. There is no understanding. There is no reading between lines. There is signal, and there is noise. Your casual typing style is mostly noise.

The Six-Band Solution

The fix is simple and mechanical. Every prompt needs 6 information bands. My sinc-prompt specification defines each one:

  1. PERSONA (n=0) — Define the expertise, perspective and tone needed for this task.
  2. CONTEXT (n=1) — Describe the situation, relevant history and operating environment.
  3. DATA (n=2) — Supply the inputs, numbers, references and facts needed to answer.
  4. CONSTRAINTS (n=3) — State the rules, boundaries, prohibitions and acceptance requirements.
  5. FORMAT (n=4) — Specify the required structure: sections, tables, code or prose.
  6. TASK (n=5) — State the objective and the work to be completed.

Key Takeaway

CONSTRAINTS states rules, limits, and exclusions to check in the output. FORMAT states the required output structure. TASK states the objective. Check all requirements in the output.

Output behavior can vary across prompts and model runs. Compare task details, model settings, sources, and output checks before attributing a difference to missing fields.

Before and After: Same Model, Different Signal

Before: Raw Prompt (1 band)

"Write a marketing strategy for my SaaS product."

For the short prompt, check whether the response assumes a target market, business model, price point, or channels that were not supplied. Verify any factual claims.

After: sinc Prompt (6 bands)

{
  "formula": "x(t) = Sigma x(nT) * sinc((t - nT) / T)",
  "T": "specification-axis",
  "fragments": [
    {
      "n": 0,
      "t": "PERSONA",
      "x": "B2B SaaS marketing strategist with 10 years experience in developer tools. Speak directly, no fluff."
    },
    {
      "n": 1,
      "t": "CONTEXT",
      "x": "Series A startup, 18 months post-launch. 340 paying customers. $42 ACV. Developer tool for API monitoring. Main competitor is Datadog."
    },
    {
      "n": 2,
      "t": "DATA",
      "x": "Current MRR: $14,280. CAC: $380. LTV: $1,260. Churn: 4.2% monthly. 78% of signups from organic search. Top 3 keywords: API monitoring, API uptime, endpoint monitoring."
    },
    {
      "n": 3,
      "t": "CONSTRAINTS",
      "x": "Budget: $8,000/month. No paid social. No enterprise sales team. Must be executable by 1 marketer. No strategies requiring >3 months to show measurable results. Prioritize channels with proven CAC under $200."
    },
    {
      "n": 4,
      "t": "FORMAT",
      "x": "3 strategies ranked by expected impact. Each strategy: 1-paragraph description, specific tactics (numbered), expected timeline, projected CAC, projected MRR impact at 90 days. Table summary at end."
    },
    {
      "n": 5,
      "t": "TASK",
      "x": "Design a 90-day marketing plan to reduce CAC from $380 to under $200 while growing MRR from $14,280 to $25,000."
    }
  ]
}

For the six-field example, check whether the response uses the supplied channel, budget, and audience data and follows the stated rules. A filled field alone does not establish factual correctness.

Same model. Same day. Same API key. The only difference was signal quality.

Use the six fields to make the specification reviewable, then check the response against the task and sources. Missing information can invite assumptions; supplying information still leaves factual, reasoning and execution errors to detect.

A prompt-review method needs its scoring rule, workload, model settings, and observed outputs before a relationship to accuracy can be assessed. Related: paper.

Every time you blame AI for hallucinating, you are saying you sent a 1-band signal into a system that needs 6. Every time you post a screenshot of a wrong answer, you are showing the world your prompt, not a failure in the model.

The question is not "why does AI hallucinate?" The real question is: why do you keep whispering into a jet engine and then complaining about the noise?

Review a prompt using six specification fields

Try sinc-LLM Free

Or install: pip install sinc-llm

// Production AI Engineering

Build AI systems that hold up in production.

sinc-LLM designs, audits, and stabilises production AI infrastructure: from vendor evaluation and cost accountability to incident controls and MCP architecture.

See what we do →
Scope and context

sinc-LLM About Blog Services

Methods, definitions, and examples for the topic discussed here. Check applicability against the task and operating conditions.