'We Don't Use AI' Will Be the Flex of 2026 — Justin Abrams & Mike Rispoli, Cause of a Kind

Show Notes

Two fifteen-year-old rock climbing buddies from Long Island made a pact in a surf lineup: build one-of-a-kind experiences for causes they cared about. That idea died. So did the loyalty app before it, and the apparel company after it. What survived was the thing nobody planned — an agency born from following opportunity instead of forcing a vision. Justin Abrams and Mike Rispoli have been failing forward together for twenty years, and Cause of a Kind is the compounding result.

The deal that let them quit their jobs was the Hospital for Special Surgery in New York, their first real foray into medical software and the moment Justin took out a $500,000 SBA loan and burned the boats. Today they build and modernize software for small and mid-sized businesses on a flat monthly model: no offshore handoffs, no surprise invoices, no equity given up, just reinvested revenue and a business that grows because it has to.

This conversation is a gut check for every founder currently drowning in shiny object syndrome. Mike has the scars of the Web3 era and sees the exact same pattern repeating with AI: companies slapping an “AI native” label on a context call to ChatGPT, then watching three competitors clone them in a month. His thesis is sharp and survivable. The magic is not AI. The magic is AI plus workflow plus deep domain knowledge — the combination that cannot be knocked off because you had to be the person on the inside to build it.

Then there is the distribution story, which is the part founders will rewatch. Cause of a Kind went from roughly 7,000 to 160,000-plus YouTube subscribers in five months by doing one unglamorous thing relentlessly: they ship every single day. No filter, no precious production cycle, just two fast-talking Long Islanders who treat their business as a media house and treat publishing as the cheapest sales conversation on earth.

Frameworks from This Episode

The Survivable AI Formula

AI alone is a commodity. Workflow plus AI is a feature. Only the third combination builds a real moat.

›AI alone gets cloned in 30 days. The barrier to copy is near zero.
›Workflow plus AI gets you a feature, not a moat. Still reproducible with effort.
›AI plus workflow plus deep domain expertise is the magic spot — you had to live the domain to build it.
›Domain expertise comes with distribution baked in: the people who trust you most are the people who most need what you built.

Follow Opportunity, Become Passionate

Cause of a Kind's entire twenty-year career arc came from saying yes to what showed up, not from executing a fixed vision.

›Start with what shows up. The market is telling you something every time someone pays for your time.
›Loyalty app, apparel, websites, software, fractional CTO — each move followed demand rather than ideology.
›Passion is more often the outcome of mastery than the input to it.
›Burning the boats (the SBA loan, quitting the jobs) works when there is already a signal worth chasing.

Experiment and Exploit

Run many small experiments to catch signal with the least effort. Kill more than you release. Some bets need years of patience before they pop.

›Keep experiments cheap and fast. The goal is to find signal, not to perfect execution.
›Kill more than you release. Most things should not survive contact with reality.
›Some bets — the podcast, the newsletter, the meetup — take years before the numbers suddenly move.
›Work from the goal backward: what is the least you can fail at to get there?

Enshittification vs. Main Streetification

Two simultaneous forces are reshaping software. One degrades quality; the other opens access. Smart founders position for the second.

›A wave of vibe-coded internal tools is coming that will break constantly and require babysitting.
›At the same time, niche professional software becomes affordable and specific enough for small businesses.
›Founders will eventually pay $300 a year for a tool that works reliably rather than maintain their own broken build.
›The opportunity is in building the reliable, specific version that Main Street businesses will pay for.

Delusional Confidence

The founders who survive long enough to compound all share a slightly unhinged commitment to mastery. Tell yourself daily you are the best in the world at your craft — and then earn it.

›Mastery has infinite room. There is no ceiling to get good at.
›'Keep your eyes on the road, otherwise you hit the barrier.' — distraction is the mechanism of most failure.
›Delusional confidence is not arrogance. It is the belief that the gap between where you are and where you are going is closable.
›Twenty years of compounding work between two people produces something that looks overnight from the outside.

Key Terms

Fractional CTO: A senior technical leader hired part time, giving startups CTO-level engineering leadership while keeping equity and payroll cost down. One of Cause of a Kind's core service offerings.

AI wrapper: A thin product that is mostly a call to a model like ChatGPT, with little defensible logic of its own. Competitors can replicate it within a month because there is nothing beneath the API call.

Non-deterministic output: AI that can return different answers to the same input. A dealbreaker for fields like finance and accounting that need identical, repeatable results every time.

Deterministic query: A use of AI that converts natural language into a fixed, repeatable result — for example, translating a user question into a precise database query that always returns the same answer for the same input.

ML model: A custom machine learning model trained on your own data, as distinct from a general-purpose LLM. More expensive and data-hungry to build, but harder to replicate because the training data is proprietary.

Vibe coding: Building software by prompting an AI tool with loose intent rather than engineering it deliberately. Great for MVPs and prototypes; produces maintenance nightmares at production scale.

Toxic productivity: Spinning the wheels all day on the back of AI tools because you can, not because it moves the business. Credited to Prime. The output looks busy; the business does not move.

Enshittification: The gradual degradation of product and software quality over time as cost-cutting and feature bloat compound. Coined by Cory Doctorow, used here to describe the wave of vibe-coded tools that will break constantly.

Startup Igniter / DISCOWEEK / WEEK2WEEK: Cause of a Kind's productized sprint offerings for taking founders from idea to prototype to deployment in compressed timeframes.

Tools from This Episode

Cause of a Kind

Bootstrapped, onshore software and modernization agency on Long Island that builds and maintains software for small and mid-sized businesses on a flat monthly model — no offshore handoffs, no surprise invoices, no equity taken.

View in Resources Visit Cause of a Kind

Q&A

What is Cause of a Kind?

Cause of a Kind is a bootstrapped, onshore software development and modernization agency on Long Island, New York, co-founded by Justin Abrams (CEO) and Mike Rispoli (CTO). They build and maintain software for small and mid-sized businesses on a flat monthly model with no offshore handoffs and no surprise invoices. They are also the hosts behind the Strictly From Nowhere podcast network.

What makes an AI product defensible in 2026?

The combination of AI plus a real workflow plus deep domain expertise. AI alone is commoditized and can be cloned within a month. Workflow plus AI buys you a feature, not a moat. Only the third combination — where you had to live the domain to build it — creates something competitors cannot replicate quickly.

Is vibe coding good enough to build a real business?

It is great for low-stakes MVPs and prototypes. The 'good enough' version becomes a maintenance nightmare at production scale. Founders eventually hit a wall on complex workflows where handing it to a professional produces a better outcome faster than continued DIY.

Why does AI struggle in finance and accounting software?

Because LLMs are non-deterministic — the same input can produce different outputs — and those fields require identical, repeatable results every time. AI can still help build the software or convert natural language into deterministic queries; it just cannot sit in the results layer.

How did Cause of a Kind grow from 7,000 to 160,000 YouTube subscribers in five months?

By shipping content every single day with no filter, treating the business as a media house, and committing to daily publication for years before the numbers moved. No production cycle, no precious editing — just relentless, consistent output from two people who had something worth saying.

When should a small business build its own internal AI tools versus hiring a professional?

Push it as far as you can yourself, but expect to hit a wall on complex workflows where the tool needs to be reliable, secure, and maintained. The question to ask is: how many hours per week am I babysitting this tool? If the answer is more than a couple, a professional likely pays for itself.

Why do agencies with the same skills charge wildly different prices?

Because the technical product is close to commoditized — the buying decision becomes emotional and relationship-driven, like bottled water. The agency that wins on price alone is rarely the one still around in five years.

Where can I learn more about Cause of a Kind?

Visit causeofakind.com. Their podcast network lives at strictlyfromnowhere.com, which includes CTO Confidential. They also run Long Island Technologists, an in-person meetup.