"Should we add AI to our business description, too?" Over the past few months, startup founders have been raising this question again and again, and there's a shared backdrop to it. Open almost any company deck or investor presentation and you'll find a line about being 'AI-powered,' offering an 'AI solution,' or 'adopting AI' — and in far more cases than not, the phrase is a buzzword wedged in rather than a description of AI actually built into the product's core. In a recent piece for Outstanding, a Korean tech media outlet, business consultant Lee Bok-yeon argues that the question itself is wrong. Before searching for an answer, he says, you have to change the question.

At first, that sounds a bit prickly. Calling it "the wrong question" can feel like it dismisses a very real anxiety founders are wrestling with. But follow the argument, and it becomes clear this isn't a case against using AI — it's a challenge to reconsider where the thinking should actually start.

The more 'AI' goes into pitch decks, the more companies look alike

In recent years, finding the word 'AI' in a startup deck has been easy. Finding a deck without it is the harder task. Whether it's a SaaS product, a commerce platform, or a content service, there's at least one line about AI. 'AI-powered recommendation algorithms,' 'AI automation,' 'AI-enhanced customer experience' — the wording varies, but the structure is the same.

Several forces drive this. Investors ask about AI. Competitors lead with AI. Job postings never leave out AI experience. The reasons differ, but the outcomes converge: a creeping fear that leaving AI out will make the company look like it's falling behind, working its way into the deck.

The problem is that in the process, these decks start to resemble one another. What problem the company is actually solving, for whom, and why this particular team — all of that fades, and the phrase 'leveraging AI' fills the space. Investors are trying to read what sets a business apart, but when everyone is using the same word, the differences become hard to find. The word AI goes in, and the company's distinctive character gets erased.

Lee's essay pins down the issue this way: "Should we do AI?" is a question about choosing a tool. But to choose a tool, you first need a problem to solve. When the tool comes before the problem, the direction of the business ends up following the fashion of the tool.

The counterargument: slap the label on first, fill it in later

There are people who clearly disagree with this view. The strongest counterargument is a pragmatic one: "Claim AI first, build it in later."

The numbers lend it some support. From 2023 through 2025, AI-related startup funding grew rapidly as a share of total venture investment worldwide, and practitioners openly say that without the AI label, it's hard to even land a first meeting. If the market wants AI, putting the label on first as a survival strategy isn't entirely irrational.

There's a second counterargument as well: "The label pulls the direction." There are plenty of teams that declared themselves AI companies and then actually built the capability. It's not uncommon to hear that committing first and moving toward it produces faster execution.

For this approach to work, though, one precondition has to hold: there has to be a 'why' behind the AI claim — a reason for which problem you're solving, and why AI is the way to solve it. Without that reason, the team starts losing momentum in the gap between what it says and what it does. The AI label might open the door at the first meeting, but at the second meeting, investors ask about the gap. Having to repeat "we're still building it" can end up costing far more than the initial advantage was worth.

For solo founders, the question arrives in a different form

This may sound like a story about startup fundraising strategy, but for solo entrepreneurs and small teams, the same structure shows up differently. Not as "Should we use AI?" but as a vague anxiety: "If I'm not properly using AI tools, am I falling behind?" That anxiety leads to guides, courses, and newsletter subscriptions, and at some point you're spending more time exploring AI tools than doing the customer work that actually needs doing.

The more limited your resources, the heavier this cost feels. Sometimes the time it takes to learn a new tool well enough to actually use it exceeds the time the tool would save.

In this situation, I'd argue that reversing the order of the questions is the much faster path. Not "Should I use AI?" but "What task am I repeatedly spending the most time on right now?" Once that question gets a concrete answer, the next questions narrow naturally: Is there a way to reduce that repetitive work? If so, does an AI tool handle it in a practical way? Is the learning cost smaller than the payoff?

With the order set this way, choosing AI tools gets much easier. Instead of feeling pressured to try everything, you can focus your exploration on the single task eating the most of your time right now. Narrow the scope to one thing — say, this month, drafting only. The narrower the scope, the smaller the learning cost, and the easier it is to see whether it's working.

One more distinction is worth making: does the AI tool make you faster at something you're already good at, or does it handle something you're bad at on your behalf? In the first case, your existing strength gets sharper. In the second, you're unlikely to build capability in that area — and when the tool changes or disappears, you're shaken along with it.

There are people who have followed every trend perfectly and found that, somehow, what their business exists for has grown blurrier. They chase the trending format, the trending keywords, the trending tools, and one day they can no longer explain what their own brand stands for. Look back, and the businesses that endure tend to be the ones that were obsessive about the problem they were solving — not the ones that were quickest to the trend.

"Should I use AI?" isn't a question of right or wrong. But the more you rush to answer it first, the less time you spend asking where you're actually starting from — and Lee's point is a proposal to flip that order. Once the problem comes into focus, the tool can be chosen afterward.