AI is fire.
It can warm your donor relationships or burn them to a cinder, and the difference is in how you hold it.
Stephen Christopher Nill, JD is deeply skeptical of AI in fundraising. He wrote Holding Fire: A Skeptic’s Framework for the Fundraiser with AI at Your Side for development professionals who share that skepticism and still want what AI makes possible: the personal attention once reserved for a handful of top prospects, extended to every donor on the list, with a human in charge of every relationship.
Also available at Amazon · Barnes & Noble · Books-A-Million · Walmart
Ordering more than one copy? Contact us
The Core Problem in Fundraising Is a Human One
Too many development professionals carry a version of the same quiet frustration. Whether a donor base numbers forty, four hundred, or four thousand, there are more relationships than any human or team of humans can personally attend to. The annual fund donors get the same bulk letter every year, and the steady givers who have quietly increased their support go unnoticed. Somewhere on the same list, a lapsed donor who might have come back never hears from anyone at the right moment.
So development shops compromise. They create segments and tiers. They write one appeal letter and mail-merge the salutation. They focus their personal energy on the top donors and hope everyone else sticks around.
Some donors do stick around, but many drift away. Donor attrition becomes the cost of a human problem that could not be solved with human hours alone, and Holding Fire is about what changed.
AI Handles the Production Work So You Can Do the Human Work
The book’s answer starts with a healthy division of labor. Give AI a donor’s complete record and it can spot trends and trace connections that a busy fundraiser would miss. It can draft a personal thank-you note, a reengagement letter, a donor briefing, or a board report in seconds. It can review an entire donor base and report which donors are trending upward, which are at risk of lapsing, which match the profile of the organization’s best major gift prospects, and which new donors are showing early signs of deep commitment.
What AI cannot do is build a real human relationship. It can prepare a fundraiser for the conversation and even draft a convincingly heartfelt note, but the conversation itself and the heart behind the note have to come from a person. AI cannot care; it can only mimic warmth. Left unchecked, it will happily substitute its judgment for yours, and the book’s first discipline is refusing to let it.
This is why the subtitle says skeptic. Stephen’s position, stated on the book’s first pages, is that it would be foolish to adopt AI in fundraising without maintaining a deep and permanent skepticism about how it is being used. The Human-Centered AI Framework at the heart of the book is built on that skepticism: five elements that keep the human at the center of every decision and every relationship, so the fire warms instead of burns.
Ten Chapters Take You from the Fire Story to Your First Monday Morning
The chapters teach the human side of the framework, and the published Specification handles the AI side; the two together are the working partnership the subtitle promises, with AI truly at your side.
A New Kind of Fire
Chapter 1 opens with the fire story and the confession behind the subtitle, then lays out the five elements of the Human-Centered AI Framework. It also coins the term relationship-building system, or RBS, as a deliberate replacement for CRM. The chapter closes with a complete Monday morning in action: a food bank development officer reviews all 412 donor records and has a prioritized briefing, with drafts attached, before the coffee cools.
The Monday Morning Strategic Review
Chapter 2 examines the weekly anchor of the whole practice: the AI reports which donors need attention and why, and the fundraiser sets the priorities and makes every call. A sidebar shows how pointing any AI assistant at the book’s published Specification turns it from a freewheeling assistant into an ally working under the framework’s rules.
Seeing the Whole Donor
Chapter 3 teaches the two lenses the book pairs for reading any donor: the donor relationship arc, which asks where this person is in the arc of engagement, and the classic qualification model of Linkage, Ability, and Interest. Together they reveal a donor’s trajectory and potential, the dimensions that gift size alone misses.
Annual Giving, Major Gifts, Stewardship, and Lapsed Donors
Chapters 4 through 7 apply the framework to the four big jobs: reimagining annual giving, running a major gifts program with AI at your side, stewardship that feels personal (including the Rule of Seven, the book’s standard of seven personal thank-you touches through varied channels before the next ask), and winning back lapsed donors. Each chapter runs the same five-step workflow against a different challenge and shows the AI’s first draft side by side with the fundraiser’s finished version.
Setting the Guardrails
Chapter 8 covers the governance work of adopting AI: the boundaries an organization draws around its practice, and the board’s role in holding them.
Ethics, Trust, and the Human Center
Chapter 9 goes deep on the ethical questions, built around three guardrails: the hierarchy that puts the donor’s interest first, the honored-or-handled test (would the donor feel honored upon seeing the process behind a communication, or handled?), and the bias-aware human review. It includes a sample AI use policy an organization can adapt as its own.
Your First Monday Morning
Chapter 10 is the deployment chapter, where the reader points an AI assistant at the published Specification and translates the book’s views into whatever RBS the organization already runs. The chapter ends with the reader’s own first Monday morning strategic review, with real donors.
A Free Practice Portfolio Lets You Learn by Doing
Talking about individualized donor journeys in the abstract is one thing; working a donor record yourself is something else entirely. So Stephen built the Practice Portfolio: a free, complete practice RBS where Maria Chen, Dr. Sarah Huang, the Whitfields, and the rest of the book’s cast wait in full donor records. Follow along as you read; every dashboard and view described in the chapters is there, working.
The Framework Behind the Book Is Free to Read
The Human-Centered AI Framework is more than the book’s chapters. It is also published in full as the Human-Centered AI Framework Operating Specification, a free, public document at charitychannel.com/framework, and it works with any major AI assistant: Claude, ChatGPT, Gemini, or the one built into an RBS.
The Specification is the framework formalized for an AI to read and follow: the division of labor, the workflow, the data discipline, the scale principle, and the ethical guardrails, in language an AI can act on directly. In the book’s terms, the Specification is what lets a fundraiser hold the fire carefully: without it, an unattended flame; with it, a hearth fire that warms the donor relationship instead of burning it down. It also speaks to the companies building AI into their fundraising platforms, so you can point your own vendor to it.
The Operating Specification is free to read and to share with attribution to CharityChannel LLC.
Your Next Monday Morning Doesn’t Have to Look Like the Last One
Picture a Monday a few weeks from now. The coffee is still warm when the briefing appears: the donors who need attention this week, each with a specific reason and a starting draft. You adjust where your personal knowledge sharpens what the AI wrote, and every gift is acknowledged within 24 hours — with a human making every call that touches a donor, and the fire held exactly where it belongs.
Also available at Amazon · Barnes & Noble · Books-A-Million · Walmart
Ordering more than one copy? Contact us