Everywhere I look, entrepreneurs are rushing to create "AI-powered" this and "AI-enhanced" that.

I’m an “AI guy” so I get it.

But…it’s bad business.

Simply adding "AI" to a business description has become the default move, as if that alone guarantees success.

We’ve been here before in the late 1990s/early 2000s when everyone added .com to their name to be an “internet company”.

Remember Pets.com?

Their sock puppet mascot got a Macy’s parade float, a super bowl advert and was featured in TIME.

All based on hype about the internet, which only had about 100 million users at the time. It all came crashing down 2 years later, along with the entire stock market.

We may be seeing something similar with “AI” being slapped on every business, without anyone knowing what that actually means. 

A few months ago I was chatting with someone who was developing an "AI-powered content management system." When I asked him what specific problem it solved that existing systems didn't, he struggled to articulate an answer. He had fallen in love with the technology, not the problem.

The harsh reality? Just adding AI to something doesn't make it a good business idea. Not even close.

What does then?

Let's get started

Business = Problem-Solving

Instead of chasing AI for its own sake, we need to return to fundamental business principles. At its core, every successful business solves a problem that someone is willing to pay to have solved. Full stop. Period. End of sentence.

This is true across all industries and price points.

Even luxury items solve problems – the need for status, exclusivity, and social signalling. A £10,000 watch solves the problem of how to demonstrate success and taste to certain people.

There's a direct relationship between your business success and

1. How painful the problem is

2. How many people have this problem

3. How effectively you solve it

Peter Drucker put it thus: "The size of the problem you solve determines the size of the opportunity." The most successful businesses tackle the most significant problems.

Let’s turn to another heavy hitter. Peter Thiel: "All happy companies are different: each one earns a monopoly by solving a unique problem. All failed companies are the same: they failed to escape competition."

This is crucial for AI entrepreneurs. Your competitive moat won't be the AI itself (which is increasingly commoditised), but the specific, painful problems you solve with it.

Hell, our customers likely won’t care that we are using AI to solve their problem. They just want the solution!

This will be our guiding principle throughout this Playbook.

The Value-Problem Connection

There's a direct relationship between the severity of a problem and how much people will pay to solve it.

Think about it this way

• A minor annoyance might be worth a few dollars to fix

• A significant time-waster could be worth hundreds

• A business-threatening issue might be worth thousands or more

This is why starting with problems rather than technologies is so crucial. When you solve a genuine, painful problem, pricing becomes much easier - the value is self-evident to your customers.

Because I’m on a roll (and believe in the Rule of Threes!) Paul Graham once said, "The best startup ideas tend to have three things in common: they're something the founders themselves want, that they themselves can build, and that few others realise are worth doing."

OK that’s the last older white dude I’ll be quoting. Promise.

That first part is key “they're something the founders themselves want” - solving a problem you personally understand gives you tremendous insight and advantage.

The Proximity Principle

One of the biggest myths in entrepreneurship is that great business ideas come from mysterious "Eureka!" moments. Research actually shows that successful founders typically build businesses in areas they're deeply familiar with.

Why? Because proximity to problems gives you

• A visceral understanding of the pain points

• Knowledge of existing solutions and their shortcomings

• Credibility with potential customers

• A built-in network for early feedback and sales

When I teach my AI Audience Accelerator I tell people to focus on educating their industry, their niche, their area of expertise. Ditto for when I license my workshop material for people to go and teach in businesses. Which businesses should they focus on? Those in their industry.

It’s a massive comparative advantage and shouldn’t be wasted. This is especially true when it comes to AI solutions. Generic AI applications rarely gain traction.

If it’s all things for all people no-one will want it. Hell they’ll just use ChatGPT most likely as it is truly all things for all people! Want to compete with OpenAI? No…me neither!

Instead if we build deeply contextualised solutions - the ones that scratch a specific itch - that’s where we can deliver real value. And make real money.

Mining Your Personal Experience

Enough fluff. Enough quotes. I had enough of that during my MBA.

Let's get practical. Your own experience is the richest, most accessible source of problem insights.

There are a tonne of methods for extracting these from you. You could just be mindful and note things down as they come to you.

But honest self-reflection can be challenging – we're often blind to our own inefficiencies or accept frustrations as "just the way things are." To overcome this, I've created several prompts that can help you extract business problems through guided introspection.

These prompts focus specifically on business activities, not personal ones. For even better results, use them with colleagues or others in your industry who might spot problems you've grown accustomed to.

Daily Friction Interview Prompt

I want you to interview me to identify business problems in my daily work that could be solved with AI. Ask me questions one at a time about:

1. Repetitive tasks I perform regularly
2. Processes that take longer than they should
3. Information I frequently need to search for or organize
4. Decisions I make repeatedly using similar criteria
5. Points of frustration in my typical workday
6. Software or tools I use that have limitations
7. Manual work I do that seems like it could be automated

For each answer I provide, ask 1-2 follow-up questions to dig deeper into the problem before moving to the next topic. After the interview, summarise the potential business problems identified, ranking them by apparent pain level and frequency.

Time Audit Analysis Prompt

I'm going to share my activities for [period of time]. For each activity, I'll include:
- What I was doing
- How long it took
- How frequently I do this task

Based on this information, please:
1. Identify tasks that appear to consume disproportionate time
2. Flag repetitive activities that might be candidates for automation
3. Detect patterns in how I spend my time
4. Suggest 3-5 potential business problems worth solving based on this time audit
5. For each problem, briefly describe how AI might help address it

Cost Center Analysis Prompt

I'll share my recent business expenses. For each expense, please:
1. Identify what problem I'm paying to solve
2. Evaluate if this solution seems efficient or optimal
3. Consider if AI could provide a better/cheaper solution
4. Rate how painful/important this problem seems (based on the amount spent)

After analysing all expenses, suggest 3-5 business problems that appear most significant based on my spending patterns, and briefly outline how AI might address them more effectively.

Decision Fatigue Identification Prompt

Help me identify decision-making processes in my work that cause friction or fatigue. Ask me questions about:
1. Decisions I make repeatedly throughout my workday
2. Choices that require gathering the same types of information
3. Decisions that follow predictable patterns or rules
4. Areas where I find myself procrastinating due to decision complexity
5. Decisions where I've made costly errors in the past

For each area I describe, probe deeper with 2-3 follow-up questions to understand the full context, information requirements, and pain points of the decision process. Then summarise the key decision-related problems that could potentially be solved with AI assistance.

These are just to get you started. Ultimately you know best what problems people in your industry face. These prompts are just ways to kick start the process of getting them out on to the page.

For now set up a Problem Bank - a note on your phone or wherever you can easily access and edit. Make this as frictionless as possible - don’t start setting up complex systems. That’s procrastination creeping in.

Also, don't worry about evaluation or filtering yet - we'll cover that in later parts. Right now this is brain dump.

What's Next?

In this Playbook I’m taking you through a comprehensive framework for finding and validating business problems worth solving with AI:

Part 1: The Problem-First Mindset - We've covered why starting with problems rather than technologies is crucial, and explored techniques for mining your personal experience (this Part).

Part 2: Problem Discovery Beyond Personal Experience - We'll expand our problem discovery to market observation and data analysis, adding even more opportunities to your Problem Bank.

Part 3: Problem Categorisation & Segmentation - You'll learn how to organise and prioritise problems to identify patterns and high-value clusters using AI-assisted analysis.

Part 4: Problem Validation & Verification - We'll explore methods to verify that problems are real, painful, and worth solving before investing in building solutions.

Part 5: Building Problem-Solving AI Businesses - Finally, we'll look at how to turn validated problems into cohesive business offerings matched to the right technical approaches.

In short we’re going to first explore and long list as many ideas as possible. Then we’re going to start the filtering and validation process to come up with the specific problems we can build a business around.

Keep Prompting,

Kyle