Soundings
Our Perspectives and Insights
AI Is Not a Solution. It’s an Accelerator.
Every leader I talk to right now is asking some version of the same question: what’s our AI strategy? Buried inside that question is a quiet, expensive assumption: that AI is a thing you adopt and a problem you solve. It isn’t. AI doesn’t fix a broken business. It makes a business move faster in whatever direction it’s already pointed, including off a cliff.
I learned this years before anyone debated AI strategy. I led the delivery team on a complete application rewrite meant to improve a working piece of enterprise software. The order came down to throw everything at it, death march style (last time I ever did that!). We shipped on time. Then nobody accepted it. We had speed and conviction from the sponsors, but never a validated direction, so we built the wrong thing, faster. Only after the failed launch did we step back, validate the direction, and rebuild. The next release landed: same team, same effort, opposite outcome.
Why the Distinction Matters
Calling AI a solution sets the wrong expectation, and the wrong expectation leads to the wrong investment. A solution closes a gap. An accelerator multiplies whatever force you already apply. Those are not the same purchase. If your business model is sound, AI compounds the advantage. If it isn’t, AI compounds the problem: faster, more confidently, and at greater scale than your old, slower mistakes ever could.
AI doesn’t fix a broken business. It makes it fail faster.
This isn’t semantics. It changes what you measure, what you fund, and what you should expect to get back.
The GLP-1 Lesson
Consider the GLP-1 weight-loss drugs everyone has been talking about. They accelerate results dramatically (real, visible, fast). But they aren’t a cure for the underlying system of diet, habits, and metabolism. Stop the drug without changing that system, and the results reverse. The drug was an accelerator, not a solution.
AI behaves the same way. It’s phenomenal leverage on top of a sound system, and a temporary illusion on top of a broken one. The moment you stop, or the moment the market catches up, the advantage evaporates if the fundamentals were never there to begin with.
Leverage Beyond Cost
Most AI conversations start and stop at cost: how many roles can we cut? That’s the smallest version of the opportunity. The real leverage is speed, reach, and decision quality: the ability to test ten ideas in the time it used to take to test one. Cost savings is a one-time floor. Capability is a compounding ceiling.
Cost savings is a one-time floor. Capability is a compounding ceiling.
But here’s the catch the cost-cutters miss. Leverage is direction-agnostic. It amplifies good decisions and bad ones with equal enthusiasm.
Speed in the Wrong Direction
I’ve written before that velocity is two things: magnitude and direction. AI is pure magnitude. It will gladly take you toward the wrong goal at a remarkable pace, and it will do it with total confidence. Peter Drucker said it best, in a line now commonly paraphrased:
“There is nothing so useless as doing efficiently that which should not be done at all.”
– Peter Drucker
AI is efficiency with the safety off. A company with a weak strategy and powerful AI doesn’t suddenly get a better strategy. It gets to be wrong faster, and at scale. So “let’s add AI” is not a plan. The plan is the direction. AI is the throttle.
The Market Will Catch Up
Today’s AI advantage feels like a moat. It isn’t. Once every competitor has the same accelerator, the speed normalizes, and you’re back to competing on the thing you were always competing on: the quality of your business model, your culture, and your decisions. AI raises the floor for everyone, which means your durable advantages, not your borrowed ones, decide whether you win after the novelty wears off.
A Framework: Accelerate the Right Things
Before you point AI at a problem, run it through five questions.
- Is the underlying system sound? Don’t accelerate something you wouldn’t want more of. Fix the model first, then add throttle.
- Are we accelerating capability or just cutting cost? Cost is a floor; capability compounds. Aim higher than the headcount math.
- Which direction are we actually pointed? Speed only helps if the heading is right. Confirm the destination before you add power.
- What happens when competitors have this too? Assume the advantage normalizes, and invest in what stays yours: culture, judgment, relationships, proprietary insight.
- Where does a human stay on the wheel? Accelerators need steering. Decide deliberately where human judgment has to remain.
The Bottom Line: Fix the Direction, Then Add Power
AI is the most powerful accelerator most organizations will ever touch. But an accelerator is only as good as the machine it’s bolted to and the hands on the wheel. Point it at a sound business and a clear direction, and it compounds your advantage. Point it at a broken one, and it compounds your problems: louder, faster, and with great confidence.
The work that matters hasn’t changed. Build a model worth scaling, choose a direction worth pursuing, and keep a human on the wheel. AI just decides how fast you get where you were already going.