Module 41Stage: Research → Create90 min

AI Fundamentals

Explain what generative AI does and doesn't do, in plain terms.

Jump to lesson:
01What Is Generative AI?02AI for Marketers03AI for Content Creation04AI-Assisted Research05Human Review
1
Lesson 1

What Is Generative AI?

Objective

Explain what generative AI does and doesn't do, in plain terms.

Teach
Generative AI predicts likely next words/pixels based on patterns in training data — it's extremely useful for drafting and pattern-matching, but it doesn't "know" facts the way a database does, so verification matters.
Real-World Example
AI can draft a caption instantly that sounds like your brand, but it might also confidently state an incorrect statistic — both are true at once.
Common Mistake
Treating AI output as automatically factual, without applying the same skepticism you'd apply to any first draft.
Practical Exercise

Ask an AI tool a factual question in your niche and verify the answer independently.

Lesson Checkpoint

Generative AI primarily works by

2
Lesson 2

AI for Marketers

Objective

Identify where AI adds the most leverage in a marketing workflow.

Teach
AI is strongest at speeding up drafts, research summaries, and repetitive variations — weakest at final judgment calls about brand fit, strategy, and truth.
Real-World Example
Using AI to draft 5 caption variations quickly, then applying human judgment to pick and refine the best one.
Common Mistake
Using AI for high-stakes final decisions (like brand strategy) instead of the fast-draft work it's best suited for.
Practical Exercise

List 3 tasks in your workflow where AI could genuinely save time, and one where it shouldn't make the final call.

Lesson Checkpoint

AI adds the most value in marketing when used for

3
Lesson 3

AI for Content Creation

Objective

Use AI to draft, not to publish unreviewed.

Teach
AI-drafted content should always pass a human review for brand voice, accuracy and appropriateness before publishing — this is exactly why KoraSpace's Brand Brain exists (Section 7.1).
Real-World Example
An AI draft that's 90% right but uses a phrase the brand would never say — caught only by human review against the brand voice.
Common Mistake
Publishing AI drafts directly without any human review step.
Practical Exercise

Draft one post with AI, then list what you'd change before publishing.

Lesson Checkpoint

AI-drafted content should

4
Lesson 4

AI-Assisted Research

Objective

Use AI to accelerate research while maintaining accuracy.

Teach
AI can summarize and synthesize information quickly, but claims that matter (stats, dates, prices) should be verified against a primary source before use — the same standard as any research.
Real-World Example
Using AI to summarize competitor positioning quickly, then checking the competitor's actual website before publishing any specific claim about them.
Common Mistake
Publishing an AI-generated statistic or claim without checking it against the original source.
Practical Exercise

Take one AI-generated claim and trace it back to (or fail to find) a primary source.

Lesson Checkpoint

AI-assisted research claims that matter should be

5
Lesson 5

Human Review

Objective

Build a lightweight but consistent human review step into any AI-assisted workflow.

Teach
A human review checklist for AI content should cover: factual accuracy, brand voice match, appropriateness, and whether it actually answers the original need — fast, but never skipped.
Real-World Example
A 2-minute review checklist applied to every AI draft before it enters the approval flow from Module 15.
Common Mistake
Treating human review as optional "if there's time" rather than a required step.
Practical Exercise

Write your own 4-point human review checklist for AI-assisted content.

Lesson Checkpoint

Human review of AI content should be

Deepen Your Knowledge
  • •Anthropic and OpenAI documentation — how generative AI models work
  • •HubSpot Academy — AI in marketing lessons
Module Checkpoint Assessment

Test Your Mastery of Module 41

5 questions · 70% passing threshold required

PreviousMod 40: Authority & Distribution
Next ModuleMod 42: Prompt Engineering