How to use this book

Who this book is for

This book was written for people who create advertising copy and content with generative AI, send them to customers, and measure performance. The primary readers are content marketers who write detail page and banner copy, brand managers who run influencer and review marketing, CRM managers responsible for text message, push, and email campaigns, and team leads who approve their deliverables.

Generative AI has made marketers faster, but it has not shifted responsibility. 'No. 1 in Sales' written by AI is also a claim of the advertising company; an AI discount text message written differently for each customer is also advertising information; and an AI-generated virtual model's "I tried it and loved it" is also a Recommendation that can deceive consumers. This book addresses the judgments practitioners must make at those boundaries: what may be used, what must be revised, to whom and when it may be sent, how to measure performance honestly, and how to bind all of this into team procedures.

A white belt (AI fundamentals) level of understanding and practical marketing experience are sufficient. Legal and statistical terms are explained each time they appear for the first time. Advertising media buying strategies, legal advice on individual cases, and ad-tech development are not covered.

Upon completing this book, you should be able to directly produce an AI copy review checklist, a context-specific sponsorship disclosure guide, a sendability determination matrix, an AI content experiment plan, and a brand review approval flowchart. The hands-on tasks in each chapter guide you to produce these deliverables one by one.

How this book is organized

The main text consists of five chapters in the same order as the exam domains. Chapters 1 and 2 cover what can be used in advertisements (substantiation of copy, disclosure of recommendations and reviews); Chapter 3 covers to whom and how those advertisements can be sent (consent to receive and customer data); Chapter 4 covers how to honestly measure the performance of sent advertisements; and Chapter 5 covers how to bind all these judgments into a team review workflow. The judgment criteria learned in earlier chapters are woven together to serve as the material for checklists and approval flows in Chapter 5.

Each chapter consists of ten sections: Learning objectives, Introduction, Key concepts, Going deeper (trade-offs and edge cases), At work (fictional case studies that even show discarded options), Common mistakes, Summary at a glance, Check the official docs, Self-check (explanations in Appendix B), and Hands-on tasks.

The appendix contains a Glossary (A), Self-check explanations (B), Exam Day Guide (C), and a list of official documents cited in this book (D). It is recommended to read in chapter order at first, and when reviewing, to start from "Summary at a glance" and "Common mistakes" of each chapter.

The exam at a glance

AI Marketing (AMK) is a Specialty course. The Specialty exam consists of 35 questions to be solved in 60 minutes, and a raw score of 75 points or higher is a pass. The free practice set consists of 10 questions and 25 minutes.

Statutory questions are based on statutes and notifications in effect at the time of the exam. For questions where the reference date changes the conclusion, the reference date is specified in the question. All companies and individuals appearing in the questions are fictional.

What the questions look like

Questions appear in five formats. Most present the situation of a fictional company first, and then ask for the most appropriate judgment or action in that situation.

A significant portion of the single-choice questions in this course are constructed as situation-flip sets. Two or three questions share the exact same five options in the same order, but because the situations differ, The correct answer varies from question to question. For example, even within the same 'AI discount text message' option set, The correct answer changes depending on whether the recipient is a recent purchasing customer, a long-dormant customer, or whether it is sent at 10:00 PM. Therefore, the habit of choosing the "most exemplary answer just by looking at the options" does not work on this exam. The factual context of the situation always determines the answer.

Distractor options are typically created in three ways. First, shifting responsibility elsewhere ("AI wrote it, so it is AI Company responsibility," "An agency sent it, so it is agency responsibility"). Second, covering substance with form ("It is sufficient to add a disclaimer at the bottom," "It is sufficient to mark it as 'AI-generated'"). Third, directly importing a principle that is correct in another situation (such as applying the nighttime exception for email to text messages). The "Common mistakes" section in each chapter organizes these patterns chapter by chapter.

Six-week study plan

This plan is designed based on 40–60 minutes per day, 5 days per week. If you have extensive practical experience, you may combine Weeks 1 and 2. At the end of each week, there must be a tangible deliverable left in your hands.

Notation and sources

Article citations. Statutory articles are written in formats such as "Article 50(3)", and Enforcement Decrees as "Enforcement Decree Article 61(2)". Articles cited without specifying the name of the statute refer to the primary statute covered in that chapter: Chapter 1 covers the Fair Labeling and Advertising Act, the first half of Chapter 3 covers the Network Act, and the second half of Chapter 3 covers the Personal Information Protection Act. When citing other laws, the name of the statute is attached.

Administrative rule citations. Public notifications and established rules of the Fair Trade Commission are abbreviated. "Operating Notification on Substantiation" refers to the "Operating Notification on Substantiation of Labeling and Advertising Content", "Designation Notification" refers to the "Notification on Designation of Types and Criteria of Unfair Labeling and Advertising Acts", "Comparative Review Guidelines" refers to the "Review Guidelines on Comparative Labeling and Advertising", "Deceptive Review Guidelines" refers to the "Review Guidelines on Deceptive Labeling and Advertising", and "Review Guidelines on Endorsements, Testimonials, etc." refers to the "Review Guidelines on Labeling and Advertising Regarding Endorsements, Testimonials, etc." Because review guidelines are criteria used by the Fair Trade Commission when reviewing illegality, even if a case corresponds to an example, it might not be illegal if there is no likelihood of misleading consumers, and it may be illegal even if not listed in the examples.

Verb force. "Shall" and "shall not" denote obligation and prohibition; "may" denotes discretion or authority; and "deemed" means that the law considers it as such. If the verb in an option differs from the original text, it is highly likely to be an incorrect answer.

Vendor names. The names of experimentation tools and AI services retain their original notation from official documentation, and prices and plan names are not covered.

Fictional organizations. All companies and individuals in the text are fictional.

Callout boxes. Note (NOTE) provides supplementary context, Tip (TIP) highlights points you must remember, Warning (WARNING) marks frequent points of failure, and Example (EXAMPLE) presents tasks to try directly.