AI Doesn't Know You: How to Get Better Results

The foundational principle most people miss when using AI

Tim Burnham, Founder & CEO

Tim Burnham

Founder & CEO

February 15, 2026

From the archive. This one is from an earlier chapter, when a lot of our work was AI training, websites and turning spreadsheets into apps. We still enjoy that work and still take some of it. These days we mostly build GTM systems for sales teams. Left up because it still reads fine and someone might find it useful.

Why Does AI Give Generic Answers?

When I first started doing AI trainings across manufacturing, insurance, hospitality, and service industries, I wrote down one line that has become central to how I teach people to use AI:

"AI doesn't know you."

Most people — because AI gives words that sound good — think that it has some intuition about them. When in reality, it's just creating good words.

How Does AI Actually Work When You Talk to It?

AI has consumed an enormous amount of text. Imagine reading every book in the world, every YouTube transcript, and every internet blog. AI is extremely "book smart," but only in a groundhog-day type way.

Every time you talk to AI, here's the formula:

Training knowledge + Your prompt = Your answer

After that answer, it forgets everything about you. The next conversation starts completely fresh.

If you understand this one thing, your results improve dramatically. When you talk to your favorite AI model like it doesn't know anything about you, its answers suddenly make more sense — and you know exactly how to make them better.

How Do You Give AI the Context It Needs?

Since AI starts from zero every time, your job is to bridge the gap between "general knowledge" and "your specific situation." Here are the most effective techniques:

Give AI a Role

This isn't intuitive, but it's one of the most powerful techniques. If you have a problem with taxes, you don't call your sales friends — you call a tax professional.

AI is your sales friend + tax professional + many other personalities combined, until you give it a specific role.

Instead of:

"I need help with my quarterly tax filing"

Try:

"You are an experienced small business tax accountant. I need help with my quarterly tax filing."

Give AI a Personality

This sounds strange, but it works. Tell AI that it's good at something, or that it pays attention to detail. You probably won't think this makes a difference, but it does.

AI is general, so you need to constrain it to get results that work better for you.

Provide Your Specific Context

Don't assume AI knows your industry, your company size, your tools, or your constraints. Spell it out:

  • Your industry: "I run a B2B manufacturing company"
  • Your scale: "We have 50 employees and $5M revenue"
  • Your tools: "We use Salesforce and QuickBooks"
  • Your constraints: "We have a limited budget and no IT team"

Share Your Standards

If you're using AI to write, tell it your voice. If you're using it for code, tell it your conventions. If you're using it for analysis, tell it what matters to you.

There are always 100 ways to solve a problem, and usually only 1-2 are the way you prefer. Tell AI your preferences first so it follows your rules.

What Happens When You Don't Give Context?

Without context, AI defaults to the most general, average answer. It's not wrong — it's just not specific to you. The answer might be technically correct but practically useless for your situation.

Expect AI to be general, unless you give it something more specific to work with.

A Simple Framework for Every AI Conversation

Before you type your next prompt, answer these four questions:

  1. Who should AI be? (Role)
  2. What does it need to know about my situation? (Context)
  3. What are my preferences and standards? (Constraints)
  4. What specific outcome do I need? (Goal)

Include those answers in your prompt, and watch the quality of responses transform.

Share

Working on something like this? Tell us what's slow and we'll tell you what we'd build.