daBongo LMS AI Training Courses

What AI Can and Cannot Do – A Practical Guide

Lesson 4: Hallucinations, Bias, and Why AI Gets Things Wrong

Lesson Objectives

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

  • Define AI hallucination in plain language with a concrete example
  • Explain why hallucinations happen at a functional level
  • Describe three types of bias that appear in AI outputs
  • Apply practical steps to reduce the impact of hallucinations and bias in their work

Lesson Content

What hallucinations actually are.

"Hallucination" in AI describes the phenomenon where a model generates specific, plausible-sounding claims that are simply wrong – invented. Not hedged, not qualified, not flagged as uncertain – just stated as fact.

Examples of hallucination:

  • Citing a journal article that does not exist with realistic author names, volume numbers, and page numbers
  • Describing the exact content of a court case that was never decided
  • Quoting a person saying something they never said
  • Providing a product specification with specific – wrong – technical numbers

Why hallucinations happen.

The language model is predicting text, not retrieving verified facts. When asked about something it does not know well, it does not say "I don't know" – it generates text that looks like an answer. The generation process does not have a "confirm this exists before outputting it" step. The result is confident-sounding text describing things that are not real.

Hallucination frequency varies by domain and task:

  • Citations and references are high-risk (AI generates plausible but often invented ones)
  • Specific figures and statistics are high-risk (specific numbers are confidently generated, often wrong)
  • Well-documented general topics are lower-risk (but not no-risk)

Practical hallucination defenses.

  • Never use AI citations without independently verifying the source exists and says what AI claims
  • For any specific figure, statistic, or technical detail you will rely on, verify against a primary source
  • Ask Claude to express uncertainty: "What are you not sure about in this response?"
  • For important factual questions, use Claude to generate a map of what to research – then research it yourself

What bias means in AI.

AI models learn from training data that reflects human writing – including the biases, perspectives, and underrepresentation in that writing. Several bias types appear in AI outputs:

Representation bias: Groups, cultures, and perspectives underrepresented in training data are less fully or accurately represented in outputs. Common subjects, Anglophone perspectives, and majority-culture viewpoints are over-represented.

Association bias: AI learns the associations present in training text. Stereotype-consistent associations may appear in generated content – in descriptions, examples chosen, roles assigned in generated scenarios.

Perspective bias: AI may present mainstream or consensus views more fully than minority, contrarian, or non-Western perspectives – not because it was designed to, but because training data coverage is uneven.

Practical bias management.

  • Recognize when AI's output may reflect under-representation rather than accuracy
  • For decisions involving diverse stakeholders, do not rely on AI output as a substitute for actually hearing from those stakeholders
  • For analysis requiring multiple perspectives, explicitly ask Claude to present contrasting or minority viewpoints alongside the main view

Practical Example

A hiring manager asks Claude to draft five sample job descriptions and notices that all five default to masculine pronouns in descriptions of the role.

He recognizes this as association bias from training data – not an intentional output, but a pattern learned from training text where masculine pronouns dominated similar content.

He adds an explicit instruction to his job description prompt: "Use gender-neutral language throughout." The outputs change immediately.

Recognizing the bias source allows him to address it directly rather than repeatedly editing each output.

Lesser-Known Tip

Hallucinations are not random – they are more likely in specific situations: when you ask about obscure topics with limited training coverage, when you ask for citations or references to specific works, when you ask about events very close to or after the training cutoff, and when the domain requires specific factual precision (technical specifications, legal citations, scientific figures). Knowing when you are in a high-hallucination-risk situation lets you apply proportionally higher verification discipline.

Safety Notes

Bias in AI outputs can have real-world consequences. AI used in hiring, performance evaluation, loan assessment, medical triage, or content moderation may reflect and amplify the biases in its training data. Before deploying AI in any decision-making context affecting real people – especially in high-stakes or regulated applications – evaluate for potential bias and ensure human oversight is in place for consequential decisions.

Practice Task

Ask Claude to answer two questions in a domain where hallucination risk is high – one where citations are likely to be needed, one involving specific technical figures. Apply the hallucination check: verify two specific claims against primary sources. Note whether any claims were incorrect. Then ask Claude to generate a scenario for a common professional situation and read the output for any association biases in how roles, genders, or groups are represented.

Completion Check

You should be able to define AI hallucination in plain language, explain why it happens, identify three types of AI bias, and describe two practical steps that reduce hallucination and bias impact on your work.

Log in and enroll to access lesson quizzes.

Scroll to Top