Đề cương khóa học
Cấp độ 1 chi tiết: chương trình học bao gồm những gì
Mỗi học phần kết nối năng lực tiếp thị thực tế với tính nguyên bản, các tuyên bố minh bạch, bảo mật thông tin khách hàng, quyền riêng tư, nhận biết chính sách nền tảng, xác minh nguồn, định hướng sáng tạo của con người, phê duyệt, khả năng kiểm toán, chuyển cấp và các quyết định có trách nhiệm.
Modern Artificial Intelligence for Marketing, Advertising and PR Professionals
What the systems sold to marketers as "AI" actually are, and why that changes how you check them. This module separates rules, machine learning and generative AI, explains how language and image models produce output, opens up the retrieval, tools and agents inside marketing products, and reads the evidence on AI use for what it does and does not show.
-
Rules, Machine Learning and Generative AI in Marketing Work
35 phút
-
How Language and Image Models Generate Marketing Output
40 phút
-
Retrieval, Tools and Agents Behind Marketing AI Products
40 phút
-
A Field Guide to Marketing AI Tools and the Evidence on Their Use
45 phút
What AI Can and Cannot Reliably Do With Claims, Facts and Creative
Why fluent output can be wrong, and what to do about it. This module covers invented sources, statistics and quotes, unsupported claims in AI search summaries, the uneven line between tasks AI helps with and tasks it harms, creative that drifts toward sameness, and the automation bias that lets errors through review.
-
Invented Sources, Statistics and Quotes
40 phút
-
The Jagged Frontier: Where AI Helps, Where It Hurts, and Why Answers Shift
40 phút
-
Homogenised Creative and the Cost of Sameness
35 phút
-
Automation Bias, Complacency and the Discipline of Doubt
40 phút
Client Confidentiality, Audience Data and Privacy Law
What client and audience information may never reach an AI tool, and why. This module covers confidential client information, where AI tools put data, state privacy thresholds and opt-outs for sale and targeted advertising, health and children's data, tracking pixels, and a decision method for putting any data into any system.
-
Client Confidences, Personal Data and Where AI Tools Put Them
40 phút
-
State Privacy Laws: Thresholds, Sale and Targeted-Advertising Opt-Outs
45 phút
-
Health Data, Children's Data and Tracking Pixels
45 phút
-
The Decision Method: Can This Data Go Into This System?
35 phút
Who Is Liable: The Agency's Own Exposure Under Advertising Law
The client is not the only party exposed when an ad deceives. This module explains the section 5 deception and unfairness tests, section 12 false advertisements, the FTC's test for agency liability, agencies as intermediaries and agents in endorsements and reviews, shared responsibility for commercial email, and the agency's duty of honesty to its own clients.
-
Deception and Unfairness: The Section 5 Tests in Agency Work
40 phút
-
The Agency Liability Test: Participation and Knew or Should Have Known
35 phút
-
Agencies as Intermediaries in Endorsements and Agents Under the Reviews Rule
40 phút
-
Shared Email Duties and Honesty to Your Own Clients
40 phút
Professional Codes, Self-Regulation and Platform Terms: What Binds Whom
Marketing has no licensing board, so every rule you meet needs a label. This module sorts member codes (PRSA, AMA), voluntary self-regulation (the National Advertising Division), trade association frameworks (IAB), advocacy pages and platform advertising policies, and shows how to tell each one from law and what it can and cannot do.
-
Sorting Authority in a Field With No Licensing Board
35 phút
-
Member Codes: The PRSA Code, Its AI Agents Advisory and the AMA Statement of Ethics
40 phút
-
Industry Self-Regulation and Trade Frameworks: The NAD and the IAB AI Disclosure Framework
40 phút
-
Platform Advertising Policies, Advocacy Pages and the Sorting Drill
35 phút
Prompt Engineering for Marketing and Communications Work
A structured way to ask AI for marketing drafts that a reviewer can check. This module builds a nine-part prompt, shows why claims and figures are supplied from the substantiation file rather than generated, sets a paste check before prompting, and covers iteration, deliberate variation, task patterns and instruction-like text hidden in pasted material.
-
The Nine-Part Marketing Prompt
40 phút
-
Supplying Claims From the File and Checking What You Paste
35 phút
-
Iterating, Varying and Task Patterns for Briefs, Headlines, Posts, Press Materials and Summaries
45 phút
-
Instruction-Like Text Hidden in Pasted Pages and Emails
30 phút
Drafting Copy and Claims From a Substantiation File
How to find every express and implied claim in AI-assisted copy, why the evidence must exist before the ad runs, and how to draft only from an approved claims file. The module also covers claims about AI itself, disclaimers that actually work, and how a competitor can challenge an ad under the Lanham Act.
-
Finding Express and Implied Claims in AI-Drafted Copy
35 phút
-
Reasonable Basis Before the Ad Runs
40 phút
-
Drafting Only From an Approved Claims File
40 phút
-
AI Washing, Disclaimers and Competitor Challenges
40 phút
Endorsements, Reviews, Influencers and Virtual Creators
What makes a message an endorsement, which connections must be disclosed and how to make a disclosure hard to miss. The module explains the federal rule on fake and incentivised reviews and the agency's place in it, how virtual creators and AI avatars are treated, and how to check whether an order or rule you have read about is still current.
-
Endorsements, Material Connections and the Intermediary's Exposure
35 phút
-
Creator Disclosures That Are Hard to Miss
35 phút
-
Fake Reviews, Incentives and the Consumer Reviews Rule
40 phút
-
Virtual Creators, AI Avatars and Checking Whether a Rule Is Still Current
40 phút
Generated Images, Video, Voice and Real People's Likeness
What the U.S. Copyright Office will and will not register when AI helped make a work, and what a federal appeals court said about human authorship. The module then covers digital replicas of real people as a matter of state law, talent contract terms for replicas, and New York's disclosure law for synthetic performers in ads.
-
Copyright Registration for AI-Assisted Creative Work
40 phút
-
Human Authorship After Thaler v. Perlmutter
35 phút
-
Digital Replicas of Real People and State Likeness Law
40 phút
-
Replica Terms in Talent Contracts and Synthetic Performer Disclosure
40 phút
Content Provenance, AI Labels and the Limits of Detection
What Content Credentials, source-type labels, watermarks and AI detectors can actually show, and what they cannot. This module teaches marketers to read provenance as a claim about origin rather than truth, to treat a missing label or a detector score as weak evidence, and to avoid the agency's own exposure from overstated verification claims.
-
What Content Credentials Can and Cannot Prove
40 phút
-
AI Labels, Source Type Values and Watermarks
35 phút
-
Why People and Machines Miss Synthetic Media
35 phút
-
AI Detector Scores, Writers and Accuracy Claims
40 phút
Building Repeatable AI-Assisted Campaign Workflows
How to turn ad hoc chatbot use into a staged, recorded process: inputs, data classification, the AI step, a claims check, a rights and disclosure check, a named reviewer and a record. The module shows how to design review so a wrong output does not become a published claim, and works through a social calendar, a press release and a product page.
-
From Chat Window to Controlled Process
40 phút
-
Designing Workflows So a Wrong Output Is Not Published
35 phút
-
Review Steps That Resist Automation Bias
35 phút
-
Worked Workflows: Social Calendar, Press Release and Product Page
45 phút
AI Agents, Chatbots and AI Voice in Customer-Facing Marketing
What happens when AI stops drafting and starts acting: agents with tools, chatbots that answer customers, and AI voices on calls. This module sets risk levels from drafting to calling, explains why a business answers for its chatbot, treats hidden instructions as a residual risk, and separates real consent and disclosure duties from ones that do not exist.
-
What an AI Agent Is and How Much Access It Needs
40 phút
-
When a Chatbot Speaks for the Brand
35 phút
-
Hidden Instructions in Pages and Messages
35 phút
-
AI Voice, Call Consent and Who Must Disclose AI
40 phút
Cybersecurity, Client Data and Vendor Evaluation
How an agency protects the client and audience data it puts into AI tools. This module shows how to read a vendor's security and privacy claims and catch changed terms, how to handle data from source to deletion, how to rank sign-in protection, what a small-agency security baseline looks like, and why data shared with ad tech needs the right contract.
-
Reading an AI Vendor's Security and Privacy Claims
35 phút
-
Client Data From Source to Deletion
40 phút
-
Sign-In Protection and a Small-Agency Security Baseline
35 phút
-
Audience Data, Ad Tech and the Contracts Behind It
35 phút
Agency AI Governance and the AI Use Policy
How an agency writes and runs an AI use policy that people actually follow. This module sets out what the policy must cover, uses the voluntary NIST frameworks as a structure, places ISO/IEC 42001 and industry disclosure frameworks as inputs rather than law, and explains where the EU AI Act may reach an agency, with its amended dates.
-
What an Agency AI Use Policy Must Cover
40 phút
-
NIST Frameworks as a Structure for Agency Governance
40 phút
-
Management System Standards and Disclosure Frameworks as Inputs
35 phút
-
Where the EU AI Act May Reach an Agency
40 phút
Practical Marketing AI Lab
Four hands-on labs with fictional clients. Find the unsupported claims and invented sources in AI-drafted launch copy, repair a creator brief and a review request, check an audience upload and a scraped page before AI touches them, and evaluate an image generator's provenance feature before drafting a short AI use policy.
-
Lab 1: Unsupported Claims and Invented Sources in Launch Copy
45 phút
-
Lab 2: Repairing a Creator Brief and a Review Request
40 phút
-
Lab 3: An Audience Upload and Untrusted Web Content
45 phút
-
Lab 4: Evaluating an Image Generator and Drafting the Policy
45 phút
Capstone: The Agency AI Implementation Plan
The level ends with a plan an agency could adopt. This module shows how to choose a low-risk first use and scope it against a baseline you measured, set controls, vendor evidence and records, read the productivity evidence by what it actually measured, and run a pilot with honest metrics, failure indicators and named stop authority.
-
Choosing and Scoping a First Agency AI Use
40 phút
-
Controls, Vendors and Records in the Plan
40 phút
-
Reading Productivity Evidence by What It Measured
40 phút
-
Pilot, Honest Metrics and Stop Rules
45 phút