— Opinion column by manufacturing expert Buckley Brinkman
AI freezes people in place. The speed and scale of this technology demand attention — but knowing who to believe, making sense of the daily flood of new information, and figuring out what to actually do — that’s the hard part.
I’m not a technology expert — I help organizations use technology more effectively. I struggle with the same information overload you do. Over time I’ve built rules of thumb — heuristics — that cut through the noise and help me move forward.
Here’s a new one I keep coming back to. It cuts through the AI haze, helps you distinguish one situation from another, and clarifies where humans fit in the picture. It sharpens my thinking — and it can sharpen yours.
HBS Professor Andy Wu built a two-by-two matrix for exactly this problem — when and how to apply AI, and where humans belong. Two dimensions define it: the type of information involved (explicit data vs. tacit knowledge) and the cost of getting it wrong (low vs. high).
The four cells look like this:
- High Error Cost/Tacit Knowledge. Think air traffic control or nuclear weapons. The stakes are too high for anything other than human judgment in the driver’s seat, with AI in a supporting role. The complexity here actually protects these jobs.
- High Error Cost/Explicit Data. AI cranks through the data, surfaces patterns no human would catch, and generates solutions to hard problems. The human in the loop validates the result. Think predictive maintenance or complex drug therapy.
- Low Error Cost/Tacit Knowledge. AI becomes your brainstorming partner—it generates options, you pick the best one. Faster thinking, broader perspective, lower cost. Good for IT troubleshooting, creative work, and anywhere good options beat narrow expertise.
- Low Error Cost/Explicit Data. AI owns this quadrant. Chatbots, routine transactions, bureaucratic processes—AI decides and executes. Humans check in only to catch feedback loop errors. If your job lives here, this capability should get your attention.
This framework helped me stop treating AI as one big undifferentiated force and start seeing each situation clearly — the risk level, the right role for people, and where to focus my energy.
It can do the same for you. Stuck on where to start with AI? Not sure you’re applying it the right way? Map your situation to one of these four quadrants. You’ll know immediately where to focus and how to bring people in.
The organizations winning with AI aren’t the ones with the biggest budgets or the deepest technical bench. They’re the ones that learned to think about it clearly. Start with this matrix. Pick one situation. Map it. Then move.
