AI isn’t a “set it and forget it” tool. To get the most out of it, you need to treat it like a team member, something you guide, train, and fine-tune over time. Over the past six months, I’ve learned that success with AI comes down to three principles: Think Partner, Feed Your World, and Iterate.

Step 1: Think Partner, Not Tool

I’ve never viewed Merlin as a tool; I see it as a partner. If I give it a vague request like “Write me a donor letter,” I get something generic. However, if I provide context such as the tone I want, the story I want to highlight, or the impact I want to emphasize, Merlin delivers something closer to what I would write myself.

The key is treating AI as if you’re onboarding a new staff member. It doesn’t know what you know, and it can’t read your mind. It needs clear instructions and direction. Once I started thinking of Merlin this way, the quality of responses improved dramatically.

Step 2: Feed AI Your World

This is where the real transformation happened. For over six months, I fed Merlin everything I could about Trusted World, our stories, tone, data, values, and even the specific words we use (and avoid).

I didn’t just correct the output; I explained why something needed to change. For example, when Merlin used the word “free,” I explained why we always say “at no cost”  because what we provide has value, and we want that value recognized. With this context, Merlin not only stopped using “free,” but it also began applying that same principle of language precision to other content.

Some examples of fine-tuning that made a big difference:

• Voice and tone: Early drafts sounded too corporate or too emotional. I taught Merlin how to strike a balance between data-driven impact and a warm, human tone.
• Mission clarity: I made sure Merlin understood that Trusted World empowers schools, police, and nonprofits; we’re not just “helping people in need.” We’re enabling others to do their best work.
• Storytelling with data: I emphasized that a great story is powerful, but a story backed by data is unstoppable. Now, data naturally shows up alongside the narratives Merlin writes.
• Family-centered impact: When Merlin wrote about a single child, I explained that one child’s need is an indicator of the entire family’s need. That changed how it frames every story.

With enough fine-tuning, Merlin started anticipating my style and preferences, just like an experienced team member who “gets it.”

Step 3: Iterate & Refine

Even with months of training, I rarely use the first draft that Merlin gives me. The magic happens in iteration. I ask for multiple variations, I adjust the tone, and I push for stronger storytelling. Each round of refinement teaches Merlin more about what I value.

The difference now is that Merlin’s first drafts are 80–90% there, because it’s been shaped by all that feedback. It’s not about perfection, it’s about collaboration. I think of Merlin as my first-draft machine, brainstorming partner, and editor rolled into one.