Most of what you hear about AI was never meant for you.
The discourse is dominated by the tech industry, venture-backed startups, large corporations, and enterprise consultants selling multi-million-dollar “transformations.” If you’re running a growing business or nonprofit with a small team and real constraints, you live in a different world. Trying to keep up with the hype cycle is more distraction than opportunity.
This post is for the people who don’t have a Chief Innovation Officer. For the teams feeling the pressure to “do something with AI” without knowing where to begin. For the ones who’ve tried ChatGPT and walked away thinking, “Neat… but now what?”
You don’t need to become an AI expert to unlock real value. But you do need to treat it like any other operational shift: grounded in reality, integrated into your systems, and built for your people.
This is the foundation. The next post will deliver tools you can use with your team to identify real opportunities and test them with purpose.
Because if you try to leap to the complex stuff without the groundwork, AI will create way more problems than it solves.
You're not behind. Everyone is still figuring this out but the wave is coming regardless of anyone's readiness for it. The ones who learn to use it now in practical ways create competitive advantage for themselves by making better decisions faster. The rest will get stuck explaining why they didn't.
🧠 AI Basics
Here's what you actually need to know to get started with AI.
Today’s buzz mostly centers around a subset of AI tools called generative AI (GenAI)—platforms like ChatGPT, Claude, or Gemini. These rely on large language models (LLMs) trained on massive datasets to predict the most likely next word in a sentence.
Think of it as a supercharged autocomplete. It’s not thinking. It’s pattern-matching.
That’s why it:
Sounds fluent, but can be wrong
Excels at writing and summarizing
Fails at math or fact-checking
You don’t need to worry about which model is best for when you’re at a low level of maturity and use. They’re all evolving quickly. They’re powerful but they need human context to be useful.
🧭 The Way to Think About AI
Too many organizations think of AI as a technology choice. Something the CIO needs to lead, something that starts with a budget or a vendor demo.
But AI, like any other tool, is really a workflow decision.
Where it adds value is in friction removal. It’s about letting your people do more of the work that matters.
Start asking:
Where are people doing repetitive, pattern-based work?
Where does inconsistency slow things down?
Where are you drowning in information without insight?
Where are teams stretched thin?
Which critical capabilities could be elevated and made stronger?
How do you free up your top talent to do more of what they’re exceptional at?
These questions don’t just point to AI—they point to leverage. And leverage is what drives growth and scale.
You don’t need a transformation roadmap. You need a win that saves two hours a week. Then another. Then another.
🤖 What AI Is Good At
Think of AI as a tireless junior assistant. It’s not deep, but it’s fast. It works best in high-volume, low-risk situations where the goal is speed, not precision. It’s best to use it to move faster on work you already understand.
Where it shines:
Ideation: Five subject lines, ten name options, or draft copy for a blog? Done.
Creative Drafting: Social posts, grant blurbs, internal memos? Get a first pass fast.
Analysis & Synthesis: Summarize survey data. Cluster meeting notes. Extract key takeaways from dense PDFs.
Lifting the Middle: AI is great at helping mid-performers get closer to senior-level outputs. It raises the floor more than it raises the ceiling.
In short: AI removes grunt work. It won’t set your strategy. But it will clear your team’s mental runway to do higher-value work.
⚠️ What AI Is Not Good At
This is where people get burned. AI is fluent, not factual. Using it in the wrong places creates confusion, cleanup, or worse. Don’t use it as a replacement for web searches, it will confidently give you the wrong answer.
Here’s where else it tends to fall short:
Precision tasks with strict rules.
Think payroll, compliance, finance workflows. “Close” isn’t good enough, and you need more complex AI tools to achieve this.Math, logic, and time.
AI guesses at numbers. It can’t reliably count, calculate, or reason about time.Facts and citations.
LLMs aren’t fact databases. They’ll make up citations and invent sources.Judgment and ethics.
AI doesn’t understand your values or impact. Use human review for anything public, sensitive, or strategic.
Bottom line: GenAI isn’t a smarter person. Don’t hand it responsibility, it doesn’t know what that means.
🧩 Common Missteps to Avoid
Here’s where well-meaning teams stall out or screw up:
Shiny object syndrome: They start with a tool, not a use case. The demo looks slick. The tool is complicated. The outcome? Confusion.
No success criteria: The team experiments but doesn’t know how to measure progress. Always ask: “How will we know this helped?”
Too big, too fast: They try to roll it out org-wide. It creates resistance or inconsistent use. Better to start small, build skill, and expand based on trust.
No change support: Even simple pilots need communication. Without clarity, people assume the worst or ignore it entirely.
The organizations seeing real impact? They start simple. They pick one internal workflow. They prove value in 30 days.
🧠 Questions to Ask With Your Team This Week
Think through these same questions from earlier alone or with your leadership:
What do we do every week that’s repetitive, but necessary?
Where do we drop the ball because someone is stretched too thin?
What’s something we’ve stopped doing because it’s too time-consuming?
Which process feels like duct tape and handoffs, not clarity and flow?
Where do we need to make our capabilities stronger?
Where can we free up our top talent to do more of their high-value work?
You’re not looking for a GenAI solution yet. You’re mapping friction.
That’s where you start.
🔧 Coming Next: GenAI Basics Field Manual
If this gave you a little clarity and context, good. We'll keep building on that. In the next Field Manual post, I’ll share:
A few guidelines to think about for security and privacy
An activity to identify which of your capabilities benefit the most from GenAI
An activity to brainstorm specific use cases to start with
A curated list of resources to deepen your knowledge
This will be the “what to do” that follows the “how to think.”
You don't need to catch up to the hype. You need to start with what's real.
- Bryan