For instructors
Teaching The Governed Enterprise
A course-fit guide for faculty evaluating The Governed Enterprise: An AI Governance Playbook for Consumer Goods as a supplementary text. Where it fits, what it covers chapter by chapter, and how to read a full chapter or request an examination copy.
Chapter 10, AI in Manufacturing and Quality, is published here in full, including its Next Steps block. It is the fastest way to judge whether it works in a classroom. Read Chapter 10.
Print or digital, at no cost. Go to the request, or write to robin.horstmann@rbcarthagepress.com.
What it is
OverviewAn AI governance playbook written for consumer goods, from food and beverage to cosmetics, pet food, and household products, and the compliance issues each one faces. General AI governance texts explain frameworks built for banks, insurers, and general enterprise. This one is written for the brand owner, the co-manufacturer, the quality organization, and the operating executive.
The book treats compliance as an operating problem, owned by the people who run the plant, the supply chain, and the systems behind them. It opens at 2:00 AM with an agentic procurement system that has committed $847,000 to a fraudulent vendor and the questions a Chief Risk Officer asks that no one can answer, then moves to a 6:00 AM call where a co-manufacturer's AI quality system has been auto-approving batches for three weeks and a consumer has found metal fragments. The second scenario carries the brand owner's recall, the retailer notifications, and a GFSI recertification audit at risk. That is the register the book works in throughout.
Appendix A states the category scope in the book's own voice: food and beverage, personal care and cosmetics, household products, pet food and nutrition, and over-the-counter health. It instructs the reader to translate food examples into their own regulatory vocabulary, and works two substitutions in full, a guaranteed analysis error at a pet food company and an active ingredient documentation gap at an OTC manufacturer.
The book is written as a supplementary text and reads at the level of decisions, evidence, and accountability. It asks who owns a system, what proof exists that it worked, and what happens at six in the morning when it did not. Students need no prior AI coursework and no technical background, and the book sits alongside a primary textbook on food safety, operations, or information systems rather than replacing one.
Where a cohort arrives with no shared AI vocabulary, Chapters 1 through 3 supply it. Chapter 1 separates traditional, generative, and agentic systems in the language of the reader's own platforms and draws the line the rest of the book works from, between content risk and action risk. Chapter 3 argues why AI governance frameworks shaped by financial services transfer poorly to a production line, then sets out five pillars synthesized from the NIST AI Risk Management Framework and ISO 42001. Those three chapters run about forty pages and make every later chapter readable on its own.
Where it fits
10 course typesFood safety and quality management
Ownership, data governance, and co-manufacturer oversight sit as a layer over the HACCP, FSMA, and GFSI material students already carry, followed by threats, risk assessment, and incident response. The question the module answers is what changes when a system decides rather than reports.
Food science and technology capstone
The manufacturing and quality chapter turns on the distinction between a modeled value and an analyzed value, which students can carry into any documentation system they meet later. Paired with the consolidated action list in Appendix A, it gives an applied project its framework.
Pet food science and technology
Chapter 10 works a case in which a pet food manufacturer's AI quality documentation system began auto-populating guaranteed analysis statements on labels from formulation targets rather than from tested values on finished product, and its Next Steps prohibit AI from populating regulated compliance documentation without qualified human verification, naming AAFCO guaranteed analysis panels alongside certificates of analysis and OTC potency records. The Appendix B entry on AAFCO states that the distinction between modeled and tested values is legally significant, and that Chapter 15 treats plant-floor AI security as a regulatory requirement under AAFCO for pet food, as it does under HACCP and FSMA for food. Third-party oversight reads directly onto a co-manufacturing relationship, where the quality system approving a batch belongs to someone else.
Supply chain and operations
Third-party oversight, agentic systems in the supply chain, and risk assessment. The supply chain chapter works as a standalone case. It opens with 23 purchase orders issued between 11 PM and 3 AM, all to one supplier, each below the $50,000 per-transaction threshold the replenishment agent was authorized to use, totaling $1,043,000, with no fraud and no malicious intent involved.
Regulatory affairs and compliance
The regulatory chapter sets out the EU AI Act's risk classification, which follows consequence of failure rather than product category, so a vision system inspecting food for contamination and a quality system on a cosmetics line both land as high-risk. It carries the penalty exposure, the FTC position that AI-generated claims face the same substantiation requirements as human-generated ones, the cosmetic-to-drug boundary, OTC monograph limits, FSMA preventive controls, and MoCRA. Its Next Steps ask students to build a jurisdictional regulatory map with a named owner and a quarterly update cycle, and to separate enacted legislation, proposed legislation, and enforcement actions as three distinct monitoring streams.
Agribusiness and food industry management
The opening foundations, the culture and change chapters, and the closing regulatory, maturity, and competitive arguments. This is the strategic frame rather than the operating detail, which suits a cohort that will manage the function rather than run it.
Personal care and cosmetic science, regulatory affairs track
Marketing and claim substantiation alongside risk assessment, which is where an AI copy tool meets a regulatory review that has not happened yet. The regulatory chapter supplies the frame, including where a wellness claim sits between FTC substantiation and an FDA drug claim.
MBA technology, ethics, or AI strategy elective
The foundations, the oversight chapters, and the closing argument that governed organizations win. Ownership and policy design are the two that most reliably produce disagreement in a seminar, since they ask who owns a system spanning procurement, quality, and finance, and what makes a policy enforceable against something that is not a person.
Information systems, IT audit, and risk
Threats, operational security, risk assessment, monitoring, and incident response map onto the control vocabulary students meet in ISACA and NIST material. The monitoring chapter separates aggregate performance from the granularity that matters, using a model at 90 percent aggregate accuracy running at 65 percent on the top five SKUs, and treats the operator override rate as governance information before it becomes a quality incident.
Professional and executive certificate programs
"For the Executive with Five Minutes," the Introduction, and Appendix A. That sequence is continuous, needs no assembly, and gives a working professional the argument and the action list without the intervening chapters.
Why it is assignable
StructureEvery chapter closes with a Next Steps block. Each one is a set of specific actions rather than a summary, which gives a ready basis for an assignment at the end of any chapter.
Appendix A consolidates all twenty-three of those blocks into one working guide, beginning at printed page 341 and preserving each chapter's subheading and lead phrase. A core subset is marked and described as the minimum viable governance foundation and the first ninety days. That marked subset is the natural scope for a semester project.
Appendix B is a glossary, which lets students from mixed backgrounds work from shared definitions. It carries full entries for the regulatory and technical vocabulary the book uses across all five product categories. A bibliography follows, giving students a route from the book into primary sources.
Named companies in the book are drawn from public reporting and cited. Other scenarios are composites that demonstrate a governance principle.
Chapter map
23 chaptersEach title links to its full description on the chapter reference page. Select a course type to light the chapters that fit it.
Part I. Foundations
Part II. Structural Requirements
Part III. Governing Specific Use Cases
Part IV. Cybersecurity and the AI Threat
Part VII. Looking Forward
Teaching materials, samples, and examination copies
A teaching packet is available to faculty on request. It carries a paste-ready syllabus insert, four worked module designs from a two-week unit to a semester-long supplementary spine, discussion questions organized by part, and eight assignment prompts labeled by level, effort, and format. Three prompts come with rubric sketches. Two put students inside a generative AI tool and then ask them to audit their own output against the book's substantiation and documentation standards. An incident tabletop comes with a facilitator sheet, timed information releases, and debrief questions.
Chapter 10, AI in Manufacturing and Quality, is published in full on this site, including its Next Steps block. Read Chapter 10.
Write to robin.horstmann@rbcarthagepress.com and name the course. It goes out by email, so no mailing address is needed.
Write to robin.horstmann@rbcarthagepress.com and name the course you are considering it for. A digital copy comes as a download link by email. A print copy ships to a mailing address, so include one if you would rather have the book in hand.
Adoption details
3 editions| Paperback | 979-8-9960620-0-3 · $29.99 |
|---|---|
| Hardcover | 979-8-9960620-2-7 · $39.99 |
| Ebook | 979-8-9960620-1-0 · $12.99 |
| Publisher | R&B Carthage Press |
| Published | August 4, 2026 |
| Extent | 469 pages, black and white throughout |
| Trade terms | A campus bookstore can order the print editions through Ingram on standard trade terms with returns accepted. |
| Online cohorts | Students in fully online or asynchronous programs can order any edition directly from the major retailers, so a campus bookstore is not required. The ebook suits a cohort spread across employers and time zones. |
Editorial review
1 outlet"A seminal and groundbreaking study, The Governed Enterprise is unreservedly recommended for personal, professional, community, and college/university Business Management & AI collections and supplemental MBA curriculum studies lists."
Midwest Book Review
About the author
Robin HorstmannRobin Horstmann is a technology leader in consumer packaged goods, currently implementing agentic AI for accounting and manufacturing and building the foundations it runs on. Her background is in the systems a company depends on, including ERP implementation and recovery, enterprise architecture, cybersecurity, and manufacturing execution and warehouse management systems. That work has run inside consumer goods for twenty years, which is why the controls get built with the system rather than bolted on after it.
She wrote the book because it did not exist. While earning the IAPP's Artificial Intelligence Governance Professional credential, she went looking for a governance reference written for consumer goods and found only frameworks built for banks, insurers, and general enterprise. She wrote the playbook afterward. Her certifications cover AI governance (AIGP), AI security (AAISM), risk and IT controls (CRISC), information security (CISM), data privacy (CDPSE), and delivery (PMP).
Full biography on the About page. She also maintains the AI Governance Regulatory Record, a tracked account of the laws, rules, and court rulings reaching consumer goods companies that use AI, reviewed on a stated cadence.