It’s Not Always About the Prompt

I love AI, but not because it gives me perfect answers on the first try.

I know some people aren’t as enthusiastic about AI as I am, but I am 100% on board with bringing this new tool into the small business world. In my own work, AI routinely helps me complete tasks in minutes instead of hours, makes me more productive, and saves me from endlessly researching obscure technical details just to solve a problem.

The more I work with AI, the more it reminds me of working with technically skilled people throughout my career. The best engineers, consultants, and problem-solvers rarely walked into a room with the answer already in-hand. They listened, formed opinions, asked questions that challenged our assumptions, and worked methodically toward a better understanding of the problem.

My collaboration with AI is becoming more like that every day.

I think one reason AI gets a bad rap is that we bring our well-honed search engine habits into an AI conversation. We ask one question, get one answer, and then judge the tool on that single interaction, good or bad. Sometimes the first answer is excellent, but the real power of AI is the ability to keep working with that answer. Done well, the collaboration between human and machine can be far more productive than a search engine ever could.

Here’s my current collaboration approach.

Don’s AI Collaboration Template

  1. Explain the problem clearly. Be sure that you are giving your AI tool all the details of the issue, and make sure that your question is as free from ambiguity as possible. (Pro tip: stop using pronouns; they can be sneakily ambiguous!)
  2. Share what you already believe to be true. Lay out your facts carefully, and be sure to talk through the areas where you have doubts and uncertainty.
  3. Ask what information is missing. Before asking for answers, ask your AI collaborator what additional information it could use to refine its responses.
  4. Gather evidence together. This was the part I didn’t understand at first, and it is essential to the collaboration. Evidence gathering is a joint effort. AI can find patterns, offer possibilities and background information, but humans bring the wisdom, experience, and perspective needed to put it all together.
  5. Challenge the conclusions. This step cuts both ways. You will want to ask your AI partner for references and ask it to challenge its own conclusions with potential alternatives. The goal is to spend enough time in conversation that you don’t leave better options on the table.
  6. Decide when it’s time to act. This step should belong to the human, and that means we’ll all reach the moment when it’s time to make a decision differently. Just be sure that the human is the one in charge of that moment.

Notice what’s missing from that list? There’s no magic prompt.

With a little practice, AI interactions will start to look less like a basic Q&A session, and more like working through a problem with a capable colleague. The tool contributes ideas, but you contribute context, experience, judgment, and decision-making. Together, the result is often better than either one of you could produce independently.

Remember, you are learning to work with a machine that can think alongside you. That takes time to get right, but it also gets better with practice. The more intentionally you collaborate, the more useful the tool becomes.

If you’re curious about where AI can create real value inside your business, book a slot on my calendar for a free, human-to-human conversation.