Session note: how non-technical leaders should think about AI in product

The Situation

AI is everywhere. Every product seems to have it. Every pitch mentions it. You're a non-technical leader trying to figure out how to think about AI in your product. Should you add AI? Should you build AI-first? What does that even mean?

The default answer is usually "add AI." But AI is a tool, not a strategy. Adding AI doesn't solve problems—it changes how you solve them. The question isn't whether to use AI. It's whether AI is the right tool for the problem you're solving.

Non-technical leaders need a way to think about AI that doesn't require technical depth. You need decision framing, not technical knowledge.

What Most Teams Try (and Why It Doesn't Work)

Most teams try to add AI because it's new and exciting, but they don't think about whether it's the right tool.

Adding AI because it's new

If you're adding AI because it's new, you're optimizing for novelty, not value. AI is a tool. Use it when it's the right tool, not because it's new.

Building AI-first without understanding the problem

If you're building AI-first without understanding the problem you're solving, you'll build AI that doesn't create value. AI-first is a solution, not a problem. Start with the problem.

Deferring to technical teams

If you're deferring AI decisions to technical teams, you're abdicating strategy. Technical teams can tell you if AI is possible, but they can't tell you if it's the right decision. That's a product decision, not a technical one.

How I Approach This in Practice

I think about AI in product by focusing on problems, not solutions. AI is a tool. Use it when it's the right tool.

Start with the problem, not the solution

What problem are you solving? Is it a problem that AI solves well? AI is good at pattern recognition, prediction, and automation. It's not good at everything. Start with the problem, then evaluate whether AI is the right tool.

Think about fit, not capability

AI can do a lot of things, but that doesn't mean it should. Think about whether AI fits the problem, not whether it can solve it. Sometimes simpler tools are better.

Avoid AI-first thinking

AI-first means starting with AI and working backward to find problems. That's backwards. Start with problems, then evaluate tools. AI might be the right tool, but it might not be.

Consider the trade-offs

AI has trade-offs. It's often less predictable, harder to debug, and more expensive than simpler solutions. Are those trade-offs worth it for your problem? Sometimes they are. Sometimes they're not.

A Real Example

A B2B SaaS company considering adding AI to their product. They were thinking AI-first—starting with AI and working backward to find problems. But when we looked at the problems they were actually solving, AI wasn't the right tool for most of them.

Instead of AI-first, we started with problems. What problems were customers actually facing? Which ones were AI good at solving? Which ones were better solved with simpler tools?

We found that AI was the right tool for one specific problem: pattern recognition in large datasets. For everything else, simpler tools were better. They added AI where it fit, not where it was new.

The outcome wasn't an AI-first product—it was a product that used AI where it made sense. They avoided the trap of adding AI everywhere, and focused on using it where it created value.

When This Matters

This approach works when:

• You're evaluating whether to add AI. You need a way to think about AI that doesn't require technical depth.
• You're being pressured to add AI. Everyone seems to have AI, and you're wondering if you should too. You need decision framing, not FOMO.
• You're building AI-first. You're starting with AI and working backward to find problems. You need to flip the thinking.

This approach doesn't work when:

• AI is clearly the right tool. If you know AI is the right tool for your problem, you don't need decision framing. Just use it.
• You're building an AI product. If your product is AI itself, that's different. This is about using AI in products, not building AI products.
• You're using this to avoid AI. If you're using decision framing to justify not using AI, that's different. Be honest about whether you're evaluating or avoiding.