To AI or Not to AI

Last week I had a call with a founder building in HR-tech. She's been in this space for years - knows the players, knows the pain points, has the kind of domain reputation that very few early-stage founders have. She'd identified a real problem, has a clear product vision, and can articulate exactly who would pay for it and why.

Her first question wasn't about product. It wasn't about go-to-market. It was: "Do I even have a chance of raising capital if this isn't an AI company? Do I need to somehow put AI into the product?"

It's not just her. Some version of this question comes up in almost every founder conversation I have these days. Founders building in health, logistics, fintech, compliance - real businesses solving real problems - and they're all quietly wondering whether "not AI" has become a liability. Whether the market has decided that if AI isn't the headline, you're irrelevant.

I don't think that's true. But I also don't think the answer is as simple as "just build a great product and the capital will come." Something has shifted, and it's worth being honest about what.

What's Actually Happening

AI companies now capture roughly half of all global venture capital. The rounds are bigger, the valuations are higher, and the headlines are bold. If you're a founder looking at this and feeling like the ground has moved - it has.

But the picture is more complicated than the headlines suggest. Strip out the AI megadeals and it turns out that US non-AI funding grew significantly year-over-year through 2025. There are growing signs of investor fatigue with the AI gold rush - bubble concerns showing up in fund manager surveys, capital starting to rotate toward companies with defensible fundamentals and real economics. Not a trickle, either - some of the larger generalist funds are visibly rebalancing toward non-AI deals with clear unit economics.

So the market hasn't shut non-AI founders out. But what it expects from them has changed.

The Wrong Question

Here's where I think founders like my HR-tech founder get stuck. "Should I add AI to my product?" mixes up two completely different things.

There's a meaningful difference between AI being your product and AI being how you build and run your company.

AI as the product means the AI capability itself is the value proposition - the model, the data, the workflow it replaces. If you're building autonomous claims processing or AI-generated legal documents, the AI is the thing. Fair enough.

But for a founder solving a real, painful, well-understood problem in HR-tech - or in logistics, or healthcare operations, or climate compliance - the product doesn't need to be "AI." The product needs to solve the problem. What changes is everything around it: how you operate, how you compete, and how you talk about value.

So the question isn't "should I be an AI company?" It's more specific than that: where does AI actually change the game for my company? And when I work through this with founders, it usually lands in three places.

Where It Lands

The first is operations. I wrote about this in "The Startup Playbook Is Dead" - one of the biggest shifts right now is what a small team can actually do. AI has compressed the gap between ambition and headcount. A three-person startup in 2026 can build, ship, and support what used to require twelve. Coding, support, outbound, customer research, finance - if you're not using AI across your operations at this point, investors will notice. They look at team size and output and they do the math.

The practical starting point is simpler than most founders think: look at where you're currently bottlenecked or where you're about to make a hire. That's probably where AI gives you the most leverage right now. A founder I work with was about to bring on a part-time content person for SEO and outbound - instead, he built an AI-assisted workflow that handles 80% of it and spends the budget on distribution. That's the kind of operational thinking investors want to see. Not "we use AI" as a checkbox, but specific decisions about where AI replaced headcount or spending, and what you did with the savings.

The second is competitive positioning. This is where it gets more interesting - and more uncomfortable. Something fundamental is shifting in how software creates value. For decades, software made people more productive. One accountant with good tools could do the work of five. AI is moving past that. In more and more workflows, it's not making the accountant faster - it's replacing the accountant.

That shift from efficiency to replacement changes what your startup is up against. If you're building HR-tech, you need a clear view of which workflows in your market are heading toward full automation - and which ones stay human because they depend on trust, context, regulation, or deep relationships. Where does your product sit relative to that line? And if a well-funded AI-native competitor entered your space tomorrow, what do you have that they can't replicate without years of your domain work?

There's a practical side to this that I think a lot of founders aren't seeing clearly yet. When AI can handle individual tasks end-to-end, it becomes possible to price work by the unit - per claim processed, per call answered, per report generated - rather than bundling everything into a salary or a SaaS license. Investors are starting to think this way. They want to hear about outcomes and what they cost, not features. If you can frame what you deliver as a measurable result - claims resolved, hours eliminated, revenue generated - and show that it's clearly better or cheaper than the alternative, that's a stronger story than any AI label. Whether there's a lot of AI under the hood or very little.

The third - sometimes - is the product itself. But even here, the question isn't "should I add AI?" It's "does AI make my product meaningfully better for the person using it?" Smart recommendations, anomaly detection, intelligent routing - features that strengthen the core value proposition - that's one thing. Gluing on a chatbot because you think the pitch deck needs it is something else.

The test I suggest to founders is straightforward: describe your product to a customer without mentioning AI. If the value proposition holds up, any AI you add is a genuine enhancement. If it falls apart without the AI label, you might have a technology looking for a problem rather than a solution looking for better tools. Investors can tell the difference - and increasingly, so can customers.

The Higher Bar

So what does all this mean when you actually sit in front of an investor?

The bar for non-AI-first companies has gone up. You need more proof points earlier. You need to demonstrate that you've thought about AI seriously - not as a product feature, but as a force that's reshaping the landscape you're entering.

In the pitch reviews and investor conversations I've been part of recently, non-AI founders keep getting some version of three questions. Where does AI genuinely help your business? Where does AI not matter to your moat - and why? And this one comes up more than you'd expect: what happens to your company if AI capability gets 10x better or 10x more expensive?

The founders who handle these well aren't the ones who tacked an LLM onto their product. They're the ones who can explain where they sit in a market that's being restructured - and why what they've built depends on something AI alone can't replicate.

Horizontal SaaS with no AI and no structural moat? That's a pitch I'd struggle to help anyone make right now. But a company built around a genuine problem, with real domain depth and a clear view of where AI helps and where it doesn't? That might turn out to be more durable than much of what's getting funded in the AI rush.

Back to My Founder

I told my HR-tech founder that she doesn't need to become an AI company. She needs to build a company that has thought seriously about AI - in how she operates, in how she competes, and in where AI features genuinely serve her users versus where they'd be cosmetic.

Her domain expertise, her network, her understanding of the problem - those are her moat. AI doesn't replace any of that. But it does change the context around it, and investors will want to see that she gets it.

Every founder I work with right now is wrestling with some version of this. And I keep coming back to something that sounds obvious but holds up: creating real value for a real customer who's willing to pay has never stopped being the point. I've believed that as a corporate executive, as a venture investor, as a founder, and as an advisor - and nothing in the current wave has changed my mind. AI changes the tools, the competitive dynamics, the investor expectations. It doesn't change that.

The question was never really "to AI or not to AI." It was always "what problem are you solving, for whom, and why should they care?" AI just raises the stakes on the answer.

I help founders navigate strategy and funding decisions when the path isn't clear. If you're there, let's talk.