The Four Ds: A Framework for Using AI With Wisdom

Our team watched a short video together this week from Anthropic's AI Fluency and Foundations course. It introduced a framework called the Four Ds — four competencies that determine whether you're actually using AI well, vs just using it.

In just five minutes, it said something that usually takes much longer to explain.

The framework comes from Rick Dakin at the Ringling College of Art and Design, and it maps four essential skills for working with AI effectively, efficiently, ethically, and safely.

Delegation — the big picture question. What are you trying to accomplish? What work should you handle yourself? Where might AI actually help? Delegation isn't about offloading tasks. It's about having a clear vision and thoughtfully dividing the work between you and the AI.

Description — clear communication with AI. There's a significant difference between saying "make me a logo" and describing your company's values, target audience, preferred colors, and style references. Description goes beyond prompts. It's about giving AI the context, goals, constraints, and tone it needs to collaborate with you well.

Discernment — evaluating what AI gives you. Are the facts accurate? Does the reasoning make sense? Does the output actually help you move forward? Discernment draws on your own expertise and requires the judgment to separate what's useful from what isn't — and to recognize when AI output needs refinement or should be set aside entirely.

Diligence — responsible use. Are you protecting sensitive data? Are you verifying accuracy before acting on it? Are you willing to stand behind the AI-assisted work you produce? Diligence means taking ownership. You're accountable for the final product, regardless of what tools you used to create it.

What our team noticed immediately is that every single D is human-first. The AI doesn't do any of the four things. You do.

As Josh put it in our discussion: "It really does include the human element throughout. It relies on each of you and your expertise to make sure that AI produces quality results for our clients."

That's the point. AI is a multiplier. As Jason on our team observed: "It multiplies what you give it — whether that's good or bad or little. If you give it very little, it can't multiply far. If you give it something bad, it can go backwards."

For ministries, this matters enormously. The theological accuracy of your content, the authenticity of your donor communications, the values embedded in your strategies — none of that can be delegated to AI. It has to come from you. What AI can do is multiply your capacity to express those things at scale.

The same morning we watched this video, our team had been demonstrating exactly these four Ds without even knowing it.

John Schwartz had been using Claude to stay accountable to client Statements of Work — telling Claude to flag him when he started doing extra work outside of scope. That's delegation and discernment working together.

Jason Skipper built a Google Ads connector that gave our team read-only access to client ad accounts directly in Claude. Within an hour, Josh was using it with a client, surfacing insights about performance max campaigns beating email revenue — data that would have taken hours to find manually. That's description and diligence: clear about what we're accessing, thoughtful about how we use it.

Rachel Slininger updated a Claude skill on her own for the first time — modifying a newsletter automation that creates HTML emails and imports them into HubSpot. She described what wasn't working, evaluated what came back, and iterated. That's the whole loop.

Anthony found something personal in the framework too: when Claude self-evaluated his usage patterns, it flagged that he rarely described the tone he was looking for — which explained why he often didn't use the responses it generated. Description matters more than most people realize.

One of our team members raised a really good question. She'd been talking with someone who uses AI to soften the tone of their professional communications — and her husband's question was: what happens when you talk to that person in real life, and you don't have AI there to help you?

It's a fair question. If AI is doing the work that builds relational skill —rather than amplifying skills you already have—you're not growing. You're outsourcing.

The Four Ds framework is a useful test for this. If you're genuinely delegating, describing, discerning, and being diligent, you're engaged throughout the process. AI is extending your capability. If any of those four D's are missing — especially discernment — AI isn't multiplying you. It's replacing you in places where you should be present.

Within an hour after our morning meeting, Josh had created a skill that everyone on the team could use to evaluate their own use of the 4D framework and provide suggestions to help them improve.

For anyone who wants to go deeper, Anthropic's AI Fluency and Foundations course is free. It's worth an hour of your time — not because you'll learn new tools, but because the framework applies to every tool, including the ones that don't exist yet.