Error and uncertainty
Use TRACE to identify fabrication, unsupported claims, omissions, contextual mismatch, invalid inference, and false precision.
Module overview
Module 4 examines hallucination, evidence, verification, AI-supported learning, scholarly research, and knowledge creation. You will identify independently verifiable claims, distinguish different kinds of error and uncertainty, evaluate source authority and independence, trace citations, and document corrections. You will also compare AI-supported learning with AI-substituted learning and design research workflows that preserve human judgment, research ethics, reproducibility, and scholarly responsibility.
Evidence and verification
Use CLAIM to identify claims, locate suitable evidence, assess source quality and independence, and record conclusions.
Learning and research
Use LEARN and RIGOUR to preserve productive struggle, authorship, methodological responsibility, and reproducibility.
Module principle: AI-generated fluency is not evidence. Important claims, citations, interpretations, and research decisions require independent human verification against authoritative sources.
Module learning outcomes
By the end of Module 4, you should be able to:
- Distinguish fabrication, factual error, unsupported claim, material omission, contextual mismatch, invalid inference, and false precision.
- Divide an AI-generated passage into independently verifiable claims.
- Apply the author-approved TRACE framework to identify errors, uncertainty, consequences, and required corrections.
- Assess source authority, relevance, currency, scope, independence, and evidential fit.
- Trace a citation or repeated claim toward its original source and identify citation laundering or false corroboration.
- Apply the author-approved CLAIM framework to document claims, evidence, assessment, correction, and remaining uncertainty.
- Distinguish AI-supported learning from AI-substituted learning and protect productive struggle, retrieval, and independent reconstruction.
- Apply the author-approved LEARN framework to a bounded learning activity.
- Use AI cautiously for research-question development, search-term generation, literature exploration, mapping, and provisional synthesis.
- Apply the author-approved RIGOUR framework to research ethics, methods, evidence, documentation, reproducibility, disclosure, and human responsibility.
Module 4 units
Hallucination, Error, and Uncertainty
Classify errors and omissions, assess possible consequences, preserve uncertainty, and apply TRACE.
Verification, Evidence, and Epistemic Responsibility
Build claim inventories, evaluate evidence, trace citations, assess source independence, and apply CLAIM.
Generative AI for Learning
Compare answer-first use with learning-preserving assistance, retrieval, reconstruction, feedback, and metacognition through LEARN.
Research and Scholarly Practice
Develop and validate research questions, searches, literature maps, syntheses, and reproducible scholarly workflows through RIGOUR.
Implementation note: replace placeholder anchors with the corresponding Moodle activity or section URLs.
Embedded Module 4 study guide
Use this two-week schedule unless Moodle specifies different dates. Begin Assignment 5 during Unit 10, build the verification record during Unit 11, and complete it before moving fully into the applied workflows of Module 5.
Week 7: Units 10 and 11
Recommended: 9–11 hours- Retrieve Module 3 safeguards.
Review information boundaries, source rights, authorship, disclosure, and human responsibility. - Read Chapter 7 and complete Unit 10.
Classify hallucinations, errors, omissions, invalid inferences, false precision, and uncertainty. - Build a claim inventory.
Break a supplied output into specific claims rather than evaluating the passage only as a whole. - Apply TRACE.
Record error type, evidence needed, possible consequence, correction, and residual uncertainty. - Read Chapter 8 and complete Unit 11.
Evaluate primary, secondary, and tertiary sources, authority, scope, currentness, and independence. - Start Assignment 5.
Select a bounded claim set and create the verification table before collecting conclusions.
Evidence to retain
- Claim inventory
- Hallucination and error classification
- TRACE analysis
- Source hierarchy
- Citation-tracing notes
- CLAIM verification table
- Assignment 1 reflection notes
Week 8: Units 12 and 13
Recommended: 9–11 hours- Read Chapter 12 and complete Unit 12.
Compare independent attempts, targeted AI assistance, retrieval, feedback, and unaided reconstruction. - Apply LEARN.
Design a learning workflow that uses AI to support rather than replace the cognitive work required by the outcome. - Read Chapter 14 and complete Unit 13.
Use AI only for bounded research support and independently validate sources, methods, findings, and citations. - Apply RIGOUR.
Document research purpose, ethics, information, methods, outputs, verification, reproducibility, authorship, and disclosure. - Complete Assignment 5.
Check every claim, source, citation, correction, uncertainty statement, and required disclosure. - Consolidate and transition.
Complete the module check and carry verified evidence into writing, workplace, and everyday workflows in Module 5.
Evidence to retain
- AI-supported learning comparison
- LEARN workflow
- Research-question and search-term record
- Verified literature map
- RIGOUR workflow
- Assignment 5 final self-review
- Updated Assignment 1 reflection
Show the core verification workflow
- Define the purpose and consequence of the information.
- Separate the output into verifiable claims.
- Identify the evidence each claim requires.
- Locate authoritative and sufficiently independent sources.
- Inspect the original source, context, method, date, and scope.
- Compare the source with the generated claim.
- Correct unsupported, misleading, or overconfident content.
- Record what remains uncertain.
- Escalate or stop when suitable evidence is unavailable.
Assessment milestones in Module 4
Assignment 5: Verification and Fact-Checking
Weight: 10%
Start: During Unit 10, Week 7
Develop: Across Units 10 and 11, then refine during Units 12 and 13
Recommended submission: After Module 4, at the end of Week 8
Official Moodle due date and time: ____________________
Assignment 1: Journal continues
Record how verification workload, source quality, uncertainty, answer-first AI use, productive struggle, and research responsibility changed your confidence and boundaries.
Recommended final submission: End of Module 6, Week 11
Official Moodle due date and time: ____________________
Assignment numbering note: Assignment 5 is intentionally completed before Assignment 4. The verification skills developed here are required for the applied workflow project in Module 5.
Show the Assignment 5 working-record checklist
- Bounded claim set and reason for selection
- Claim inventory
- Error and uncertainty classification
- Evidence requirements
- Source authority, currency, scope, and independence
- Original-source or citation tracing
- Verification findings
- Corrections and justified confidence
- Remaining uncertainty
- AI-use documentation and disclosure required by the assignment
Core learning activities
| Activity | Purpose | Product to retain | Connection |
|---|---|---|---|
| Claim inventory | Convert a fluent passage into independently verifiable claims. | Claim table | Assignments 5 and 6 |
| Error classification | Distinguish fabrication, unsupported claims, omissions, mismatch, and invalid inference. | TRACE audit | Assignments 1, 4, 5, and 6 |
| Reference audit | Verify titles, authors, dates, identifiers, methods, and attributed findings. | Verified reference record | Assignments 5 and 6 |
| Citation tracing | Trace a repeated claim to the original source and assess independence. | Citation chain | Assignment 5 |
| TRACE and CLAIM table | Document claims, evidence, findings, corrections, and uncertainty. | Verification table | Assignment 5 |
| Learning experiment | Compare answer-first use with independent attempt, targeted help, and reconstruction. | LEARN comparison | Assignments 1 and 6 |
| Research validation workshop | Generate search ideas, then validate sources through scholarly or authoritative repositories. | RIGOUR record | Assignments 5 and 6 |
Live AI, supplied-output, and non-platform options
Option A: Approved AI
Use an approved system with synthetic, public, supplied, or explicitly authorized content. Verify all consequential claims independently.
Option B: Supplied materials
Use supplied passages, references, source excerpts, outputs, and learning or research records.
Option C: Non-AI analysis
Complete the same claim analysis, source verification, learning design, and research workflow without an AI platform.
All options must assess the same verification, learning, research, and human-responsibility outcomes.
Frameworks introduced in Module 4
TRACECLAIMLEARNRIGOURSource independenceCitation tracingReproducibility
TRACE
Use the exact author-approved textbook expansion for structured analysis of errors, omissions, uncertainty, consequences, corrections, and escalation.
CLAIM
Use the author-approved framework to identify claims, define evidence needs, assess sources, record findings, and preserve uncertainty.
LEARN
Use the author-approved framework to protect independent effort, targeted assistance, retrieval, feedback, reconstruction, and metacognition.
RIGOUR
Use the author-approved framework for research purpose, ethics, information, methods, source validation, reproducibility, authorship, and disclosure.
Responsible-use boundaries, accessibility, and support
Boundaries
- Do not treat generated references or quotations as authentic until verified.
- Do not upload unpublished research, participant information, student work, confidential records, or proprietary manuscripts without authority.
- Do not use AI to replace assessed learning, scholarly judgment, research ethics review, or accountable authorship.
- Do not claim certainty beyond the evidence.
- Stop when original evidence cannot be located or suitable verification is unavailable.
Support and escalation
- Use library services for database searching, source tracing, and citation support.
- Contact the course coordinator for assignment scope, permitted AI use, and academic-integrity questions.
- Contact research ethics, copyright, privacy, accessibility, or disciplinary specialists where applicable.
- Use supplied alternatives whenever live AI use is inaccessible, unsuitable, or unwanted.