daBongo LMS AI Training Courses

← Back to Course

Getting Started with ChatGPT

Lesson 3: Learning with ChatGPT – and Knowing When Not to Trust It

Lesson Objectives

By the end of this lesson, students should be able to:

  • Use ChatGPT effectively as a learning partner and explainer
  • Explain why ChatGPT can produce wrong answers confidently
  • Apply the four-tier verification framework to any ChatGPT output
  • Describe the specific limitations that make verification essential

Lesson Content

ChatGPT as a learning partner.

One of ChatGPT's most valuable capabilities is as a personalized explainer. Unlike a search engine that returns existing articles, ChatGPT can:

  • Explain concepts at exactly your level of background knowledge
  • Answer the specific follow-up question that occurred to you
  • Use analogies from your field or experience
  • Test your understanding with tailored questions
  • Adjust the explanation when it misses the mark

To use ChatGPT as a learning partner effectively:

"I want to understand [concept]. I have a background in [your background] but no experience with [new topic]. Explain it in a way that builds from what I already know. After you explain, give me three questions to test my understanding – ask them one at a time."

Why ChatGPT can be confidently wrong.

Understanding this is the most important safety concept for new users.

ChatGPT generates text that is statistically likely to follow from what came before – it is trained to produce plausible, coherent, well-organized language. But "plausible and well-organized" is not the same as "accurate." ChatGPT can produce:

  • Hallucinated facts: Statistics, citations, quotes, and details that sound real but are fabricated
  • Outdated information: Its training data has a knowledge cutoff – events after that date are not in its knowledge base
  • Confident errors: Mistakes expressed with the same tone and structure as accurate information

This is the most important thing to internalize: ChatGPT's confidence level does not indicate accuracy. A confidently stated hallucination reads exactly like a confidently stated fact.

The four-tier verification framework.

Match your verification effort to the stakes of the decision:

Tier 1 – Low stakes (plausibility check): Mistakes are easily corrected with little consequence. *Examples: brainstorming, casual learning, organizing your thoughts, drafting internal notes.* Verify: Does this seem plausible? Does it align with what you know? A quick sanity check is sufficient.

Tier 2 – Medium stakes (spot-check important claims): Mistakes could cause meaningful cost, wasted time, or embarrassment. *Examples: client-facing documents, information you will share publicly, research summaries you will act on.* Verify: Check the most important facts and claims against authoritative sources. Do not accept statistics, citations, or specific claims at face value.

Tier 3 – High stakes (verify all claims): Mistakes could cause significant professional, financial, or reputational harm. *Examples: financial analysis, legal documents, medical information, information being presented to leadership.* Verify: Every material claim verified with an authoritative source. Do not act on ChatGPT's output without formal verification.

Tier 4 – Non-negotiable (professional review required): Mistakes could cause irreversible harm – legal, medical, financial, safety-critical. *Examples: medical treatment decisions, legal filings, financial investment advice, safety engineering.* Verify: Licensed professional review is required regardless of how authoritative ChatGPT's response appears.

Applying the framework.

Before using any ChatGPT output, ask: "What is the worst realistic outcome if this is wrong?" Your answer tells you which tier applies.

Practical Example

A student asks ChatGPT to explain the components of a business plan. She is preparing for an entrepreneurship class. The explanation is excellent – organized, clear, and covers all the main elements. Tier 1 applies: this is learning for a class, the stakes are low, and the explanation aligns with what she knows from her business textbook. She uses it confidently.

Later, she asks ChatGPT to give her statistics on startup success rates for a class presentation. Tier 2 applies: she will be presenting this publicly to her class. She spot-checks: two of the four statistics ChatGPT provided have no source she can verify, and one contradicts a more recent study she finds directly. She uses only the statistic she can verify from an authoritative source.

Same tool, same day – different verification effort because different stakes.

Safety Notes

Never ask ChatGPT for medical, legal, or financial advice and act on it without professional consultation. These are Tier 4 situations – the potential for harm is too significant and the professional judgment required is too specialized for an AI assistant, regardless of how expert ChatGPT's response appears. Always escalate to licensed professionals for decisions in these domains.

Jamie Practice Lab

Before moving to the quiz, complete this short applied exercise:

  1. Write one realistic ChatGPT prompt that applies the main idea from lesson 3 to your own work, learning, or daily life.
  2. Add one safety or verification step you would take before acting on ChatGPT's response.
  3. Revise the prompt once to include clearer context, constraints, or success criteria.

Instructor check: A strong answer should show practical use, human review, and awareness that ChatGPT output is assistance – not automatic truth or professional advice.

Added Quiz Enhancement

question_id: auto-enhancement-lesson-3-qjamie001 question_type: short_answer difficulty: applied question: Write one prompt you could use after this lesson, then name one verification or human-review step you would apply before relying on the result. correct_answer: Answers will vary; a strong answer includes a clear task, relevant context, at least one constraint or desired format, and a realistic verification or human-review step based on the stakes of the task. answer_explanation: This applied question checks whether the student can transfer the lesson into real use while maintaining responsible AI habits.

Lesson Quiz

Log in and enroll to take this lesson quiz.

← Back to Course

Scroll to Top