The New Rules of Building

A playbook for founders in the age of AI

I had the privilege of speaking at the Babson College Global Entrepreneurial Leadership (GEL) Forum last week, and the energy in the room was electric. We’re living through a period of profound change, and the students and aspiring founders I spoke with are acutely aware that the ground is shifting beneath their feet. The old playbooks are being thrown out, and a new set of rules is being written in real-time.

This post is an attempt to capture the core ideas from that talk. It’s a playbook for anyone looking to build something meaningful in 2026 and beyond. The central thesis is simple: the collapse of old advantages has created the single greatest opportunity for a new generation of founders in history.

The Great Unbundling

For decades, the barriers to starting a company were high and well-defined. You needed capital for servers, a sales team to reach customers, and a large engineering team to build and maintain complex codebases. These weren’t just operational hurdles; they were the moats that protected incumbents.

Today, those moats are being drained.

•Capital? Cloud infrastructure and pay-as-you-go APIs have turned massive upfront capital expenditures into manageable operating expenses.

•Distribution? App stores, social platforms, and product-led growth have given small teams the ability to reach millions of users without a traditional sales force.

•Code? Open-source libraries, low-code platforms, and now, generative AI, have dramatically lowered the barrier to building sophisticated software.

This isn’t a gradual evolution. It’s a seismic shift. The very things that made incumbents strong are becoming sources of weakness. Their rigid processes, their reliance on established patterns, and their sheer organizational inertia make it nearly impossible for them to compete with the speed and agility of a small, focused team.

As Clayton Christensen famously observed, the things that make a company good at its existing business often make it bad at competing for the disruption. We are living through the most dramatic validation of that theory in modern history.

The Gamma Playbook: A Case Study in the New Rules

If you want to understand what the new rules look like in practice, there’s no better case study than Gamma. In late 2022, Gamma was a two-year-old startup building a presentation tool. They had 60,000 users and less than a year of runway. They were, by all accounts, on the verge of becoming another forgotten startup.

Then, in November 2022, ChatGPT launched.

Most companies spent the next year debating whether generative AI was a fad. They formed committees. They wrote white papers. Gamma did something different. In December 2022, they bet the entire company on AI. They spent three months rebuilding their product from the ground up around a generative AI core.

They relaunched in March 2023. The result was explosive.

•3 million users in the first three months.

•Cash-flow positive within months of the relaunch.

•As of late 2025, 70 million users, $100M+ in ARR, and a $2.1 billion valuation.

All with a team of around 50 people.

Let that sink in: 70 million users. 50 employees. Do the math. The relationship between headcount and impact has been permanently broken.

Gamma didn’t predict the AI wave. They just moved faster than everyone else once they saw it. Their story is a playbook for the new era: decisiveness, speed, and a willingness to abandon old assumptions are the new competitive advantages.

The New Questions Every Founder Must Ask

The questions founders ask reveal the era they’re building in. For the past decade, the questions have been predictable:

•How do I find a technical co-founder?

•How much should I raise in my seed round?

•What’s my go-to-market strategy?

Today, the questions are different:

•Do I even need a technical co-founder? When one person can spin up a full-stack application with a few prompts, the traditional division of labor between “business” and “product” starts to dissolve.

•Can I bootstrap to profitability first? The explosion of million-dollar, one-person businesses (over 117,000 in 2023, up nearly 4x in a decade) proves that venture capital is a choice, not a necessity.

•Is my product AI-native or just AI-enhanced? This is the most important question of all. If AI stopped working tomorrow, would you have a slower, less efficient business, or would you have no business at all? The former is a feature. The latter is a moat.

Your Moat in a World of Shared Models

When every founder has access to the same APIs, the traditional software moat of proprietary technology disappears. Your defensibility no longer comes from the model itself, but from what you build around it. Here are the six moats that matter now:

1.Proprietary Data: If your product generates or ingests data that no one else has, your model gets better in ways competitors can’t replicate. The model is a commodity; the data is the asset.

2.Network Effects: If the product becomes more valuable as more people use it, you create a structural advantage that compounds with scale.

3.Workflow Entrenchment: Deep integration into a team’s processes and data raises switching costs, even if a competitor offers a marginally better model.

4.Brand Trust in High-Stakes Domains: In fields like healthcare, legal, and finance, customers will not switch to an unknown AI tool for a marginal benchmark improvement. First-mover trust is durable.

5.Distribution: A proprietary channel, an existing user base, or an institutional partnership lets you reach customers faster and cheaper than anyone building from zero.

6.Founder Domain Expertise: Deep industry knowledge creates an insight, credibility, and hiring advantage that a generalist competitor cannot replicate quickly.

The strongest companies are building two or three of these from day one.

The Only Rule That Never Changes: Start

Everything I’ve written so far is noise until you do one thing: start.

The biggest mistake founders make is waiting for the perfect plan, the perfect team, or the perfect market signal. The optimization in 2026 is not for certainty; it’s for learning speed. Ship something. Get real signal. Probably pivot. Repeat.

And when you seek that signal, understand the difference between someone saying “I love this idea” over coffee and someone actually signing up, filling out a form, or handing you a credit card. The former is noise. The latter is signal. Your goal before you raise a dollar is to get 10 people to do something, not just say something.

The best pitch isn’t a 50-slide deck.

The best pitch is: “I have customers. Do you want in?”

Everything else is just a distraction.

This post was adapted from a talk I gave at the Babson College Global Entrepreneurial Leadership (GEL) Forum in February 2026. Thanks to the organizers and students from Chris Payton’s class for the incredible energy and sharp questions.