Your board asked about AI at the last meeting. Two of your competitors put out press releases about AI features last quarter. Your VP of Engineering wants to budget for AI tools. An enterprise customer asked about your AI roadmap in a renewal conversation. And you’re sitting in a leadership team meeting where everyone is waiting for you to have a point of view.
You’re not sure if you’re genuinely behind, or if everyone is running toward the same fog together and nobody wants to say it out loud.
Here is how to tell the difference — and what an actual AI strategy looks like versus what gets sold as one.
The Question You’re Actually Trying to Answer
When leadership teams say “we need an AI strategy,” they usually mean one of three different things:
They mean: our competitors are announcing AI features and we don’t want to look like we’re standing still. This is a legitimate concern but it’s a marketing problem, not a technology problem. The answer is not an AI strategy. The answer is an honest assessment of what your competitors are actually shipping versus what they’re announcing, and a decision about whether you need to respond now or in six months.
They mean: AI tools might change how our team works and we don’t have a policy. This is an HR and security problem. It’s important and it needs to be addressed, but it’s also not an AI strategy. It’s a set of guidelines for using third-party AI tools responsibly.
They mean: AI might fundamentally change what our customers need from us, or how we can deliver our product, and we haven’t thought carefully about that yet. This is an actual strategic question, and it’s the one worth spending serious time on.
Most companies conflate all three. They call the whole thing an “AI strategy” and end up with a document that’s partially marketing, partially policy, and partially real thinking — without being enough of any one of them to actually drive decisions.
What Real AI Strategy Looks Like
A real AI strategy starts with your unit economics and your customers, not with AI.
The right first question is: where does labor, time, or error rate limit what we can deliver or what we can charge? Those are the places where AI can move the needle in a way that changes the business.
If you’re a services business billing by the hour and AI cuts the time to do work by 40%, that’s a fundamental business model question, not a technology question. Do you pass the savings to clients and compete on price? Do you keep the margin and invest in scope? Do you use the capacity to serve more clients? The AI is almost incidental — the strategic question is about what your business model looks like in a world where the unit of work changes.
If you’re a SaaS company and your customers currently do something manually that you could automate for them inside your product, that’s a product question. It might also be a competitive moat question if you can do it in a way that your competitors can’t easily replicate.
If you’re a data-rich company — you have years of proprietary customer behavior, transaction history, or domain-specific data — that data might be the foundation for a capability that competitors can’t buy off the shelf. That’s worth thinking about carefully.
Start there. Start with the actual business levers. Then work backward to the technology.
What Gets Sold as AI Strategy
AI theater is what you get when the strategy process starts with the technology instead of the business.
It looks like this: your team evaluates three AI vendors, picks one, announces a partnership, issues a press release about “AI-powered” features, and ships something that uses AI but doesn’t change the outcome for the customer or the economics for the company. The product team learned some new tools. The announcement generated some LinkedIn engagement. The substance moved nothing.
This is not a criticism of the people involved. It’s what happens when the pressure to have an AI narrative drives decision-making faster than real strategic thinking can happen.
The way to detect AI theater in your own organization is to ask: if this AI initiative works exactly as planned, what business metric changes and by how much? If you can’t answer that question with a number and a rationale, you don’t have a strategy. You have a project.
The Practical Starting Point
If you’re a CEO who needs to get oriented, here’s what the first 30 days should look like.
First: audit what’s already happening. Your engineers are almost certainly already using AI coding tools. Your customer success team may be using AI for drafting. Your marketing team may be using it for content. Before you set strategy, understand the current state — what’s being used, by whom, with what safeguards. This is your baseline.
Second: talk to your three best customers about their workflows. Ask them what work is tedious, slow, or error-prone for them, and whether they’ve started experimenting with AI tools to address it. What they tell you will be more valuable than any market research report on AI adoption.
Third: have an honest conversation with your engineering team about build versus buy. For most companies, the right AI strategy involves integrating existing models and platforms rather than building proprietary AI. But that’s a decision that requires understanding your data, your use cases, and your competitive context. It should not be assumed in either direction.
Fourth: make one small, concrete bet rather than a large, ambiguous initiative. Pick one workflow, customer problem, or internal process where AI could change the outcome in a measurable way. Run it as a real project with a real business case. The organizational learning from one real experiment is worth more than a strategy document of any length.
The Board Question
When the board asks about your AI strategy, what they usually want to know is: are you paying attention, do you have a point of view, and are you making thoughtful decisions rather than either ignoring it or chasing it recklessly?
You don’t need a 40-page roadmap to answer that question well. You need a clear understanding of where AI does and doesn’t move the needle for your business, and a concrete example of something you’re testing.
That’s a harder answer to give than a deck full of AI logos and announced partnerships, but it’s the one that holds up under questions.
If you’re a CEO trying to figure out whether you’re genuinely behind on AI or being sold to, and you want a straight read on where AI realistically moves the needle for a company in your situation, a 15-minute call with Christopher can help you get oriented. He works with companies on AI strategy that starts with the business problem, not the vendor pitch.