The DIY Tax and what building AI yourself really costs

The cash price of standing up AI yourself is the cheap part. The bill is the founder hours, and the founder turned into the bottleneck.

Technology
By Mark Choudhari · Jun 3, 2026 · 6 min read

AI Economic Growth
Made with Works

TL;DR

The DIY Tax is what a founder pays by becoming the AI implementer for their own business. The cash is small, a few thousand dollars, but it costs 60 to 120 hours of founder time, and far more once you price those hours honestly. The real bill is not the dollars. It is the founder turned into the bottleneck the rest of the business waits on.


In this article

Building AI into your business yourself looks like the responsible choice. The cash outlay is small, you keep control, and you learn how it all works on the way. Then the weeks start disappearing into reading docs, wiring tools together, and fixing whatever broke when you changed one setting. That time is the second hidden cost of AI, and almost no one puts it on the books. It is the second of seven taxes in the AI Tax, the real price of AI for a founder-led business once you count everything past the cash line. The first tax was the time you lose choosing tools. This one is the time you lose standing them up.

Should I set up AI tools myself or get help

Start with the honest number. Setting up AI yourself runs about $1,500 to $3,000 in cash plus 60 to 120 hours of the founder’s own time. The cash is the part you see. The hours are the part that costs you, and they come out of the scarcest resource in the business, which is the time of the person running it.

So the real decision is not “spend a little or spend a lot.” It is “whose time does this take.” When you do it yourself, it takes yours, the same time that should be going into customers, the team, and the work that grows the top line. That is the trade most DIY guides never price, because the cash looks so reasonable that the hours feel free. They are not free. They are the most expensive hours in the company.

Setting up AI yourself runs about $1,500 to $3,000 in cash plus 60 to 120 hours of the founder’s own time.
The Forest View, 2026

There is a fair counter-argument, and it is worth conceding. Doing it yourself teaches you how AI actually works, and that knowledge is genuinely useful. The catch is that learning is a one-time gain. The DIY Tax is the recurring cost of being the one who builds it, fixes it, and rebuilds it every time a tool changes underneath you. You can keep the lesson and still hand off the labor.

Why does setting up AI take so long

Because the slow part is not the AI, it is the integration. Connecting AI to the data and tools you already run, keeping the output consistent, and building a workflow the team will actually use needs skills most growing businesses do not have sitting around internally. When that skill is not on staff, the founder absorbs it, and the founder is learning it in real time, which is exactly why an afternoon turns into a week.

This is the part that quietly breaks the timeline. Buying a tool is fast. Making it work with the fifteen other things your business depends on is not, and it is the part nobody estimates honestly up front. The biggest delays come from data quality and readiness, not the technology, which is exactly the work that lands on the founder. We go deeper on this specific cost in the Integration Tax, because connecting the pieces is its own tax on top of building them.

Priced at the founder’s real rate, the slowness is not abstract. At $100 an hour, those 60 to 120 hours make the true cost roughly $6,000 to $12,000. State the assumption plainly: that figure assumes your time is worth $100 an hour, and most founders running a business would say theirs is worth more. Either way, the cheap option stopped being cheap somewhere around hour forty.

What does it really take to implement AI in a business

It takes more than the build. It takes the maintenance, and the maintenance is where DIY actually fails, because the time spent debugging and keeping it running can quietly exceed the time the AI saved you. The failure is rarely a dramatic crash. It is the ordinary afternoon you lose hours to a technical issue a specialist would have cleared in minutes, while the rest of the business waits on you to finish. That is the founder-as-bottleneck, and it is the real shape of the DIY Tax once the initial build is done.

The maintenance is where DIY actually fails: the time spent keeping it running can quietly exceed the time the AI saved.
DEV Community (AWS), The Hidden Cost of AI Coding, 2026

Think of it in three layers. There are the tools at the bottom, the work they do in the middle, and the business results at the top. Standing up AI yourself keeps you stuck at the bottom layer, building and fixing the plumbing, while the results at the top wait on you to climb back out. The most expensive version of this is the one where the founder never climbs out, because every change to a tool pulls them back down to rebuild.

That is the difference between learning AI and being trapped as its implementer. The first is a phase. The second is a job you accidentally gave yourself, and it is the one that taxes everything else you are supposed to be doing.

What a system that builds it for you looks like

The way out is not a better tutorial or a faster afternoon. It is to stop being the person who has to stand it up at all. Any real answer has to clear a clear bar: it has to build the workflow against your business rather than from a blank page, run the work across the tools you already use, and absorb the upkeep so a tool change does not pull you back down to rebuild. Your question changes from “how do I implement this” to “what do I want done.”

JynAI built Works, an AI Business OS, to clear exactly that bar. Here is the fit, plainly.

  • Pain: the build starts from a blank page and your own learning curve.
    Business-Aware Setup reads your LinkedIn, site, and files into a workspace that arrives already understanding the business, so there is nothing to stand up from scratch.
    Gain: the weeks you would have spent wiring it never get spent.

  • Pain: the work stalls because nobody on staff can connect the tools.
    Work That Actually Ships runs in three modes, Strategy to plan, Action to execute across the tools you already use, Automation to run hands-free, so the workflow finishes instead of waiting on you to build the next step.
    Gain: the workflow closes on its own instead of parking on your desk.

  • Pain: every tool change pulls you back down to rebuild.
    The Workflow Library and the model pool update underneath you, so when a new model ships it joins the pool and your existing work starts using it without you touching a thing, and your setup does not move.
    Gain: the rebuild that used to land on you every few months stops being your job.

  • Pain: you do not have a specialist on staff to keep it running.
    The Work Runs at the autonomy level you set, from approve-every-step to hands-free, so the upkeep is the system’s, not an afternoon stolen from your week.
    Gain: the founder stops being the thing the business waits on.

The affordability is the part that makes this honest rather than aspirational. The full capability set is available at the $49 tier, not behind an enterprise contract, so the founder who would have spent 60 to 120 hours building can reach the same ground without becoming the implementer to afford it.

And we are not theorizing. We were the implementers first. Machintel spent two years standing AI up and re-standing it up, the same hours that should have gone into running the business, before we built the layer that took the building off our plate. Once it was off, six teams were running on it inside ninety days. We are biased about our own product, of course. The argument under it does not need us: if the expensive part of DIY was always the founder’s weeks and not the cash, then a faster afternoon was never going to be the answer. Taking the building off the founder’s plate is.

The cash price of building AI yourself is the cheap part. Your weeks are the bill. The next time DIY looks like the responsible choice, price the hours before you price the tools, and ask whether the thing you are about to build is your business or just the scaffolding you will keep rebuilding around it.

Stop building yourself. Sign up for early access. Or size your own bill first with the AI Tax Calculator.

Common Questions

Should I set up AI tools myself or get help?

It depends on whose time you are willing to spend, which is the part most DIY guides never price. The cash to do it yourself is small, about $1,500 to $3,000, but it costs 60 to 120 hours of the founder’s own time, the scarcest resource in the business. If you do it yourself, keep the learning and plan to hand off the labor, because the labor is the part that recurs.

Why does setting up AI take so long?

Because the slow part is not the AI, it is the integration. Connecting AI to the tools and data you already run, keeping the output consistent, and building a workflow the team will use takes skills most growing businesses do not have on staff, and the biggest delays come from data readiness rather than the technology. The founder absorbs that work and learns it in real time, which is why an afternoon turns into a week.

What does it really take to implement AI in a business?

More than the build. It takes the maintenance, and the maintenance is where DIY quietly fails, because the time spent debugging and keeping it running can exceed the time the AI saved. The real cost is not a crash, it is the ordinary afternoon you lose to a problem a specialist would clear in minutes, while the business waits on you.

How much does it actually cost to build AI yourself?

The cash is roughly $1,500 to $3,000. The honest number adds 60 to 120 hours of founder time, which at $100 an hour puts the true cost closer to $6,000 to $12,000, and most founders would say their hour is worth more than that. The cash was never the expensive part.

What is the DIY Tax?

It is the second of the seven taxes in the AI Tax, the real price of AI for a founder-led business once you count past the cash line. The DIY Tax is what a founder pays by becoming the AI implementer: the hours to build it, the hours to fix it, and the founder turned into the bottleneck the rest of the business waits on.

Recent Posts
Getting AI working is one cost. Keeping it working is the tax.
The Discovery Tax and what choosing AI really costs
Integration Tax
The Integration Tax and why your AI keeps breaking
AI Economic Growth
Why your team never reaches the AI payoff

The DIY Tax and what building AI yourself really costs

The cash price of standing up AI yourself is the cheap part. The bill is the founder hours, and the founder turned into the bottleneck.

Technology
By Mark Choudhari · Jun 3, 2026 · 6 min read

AI Economic Growth
Made with Works

TL;DR

The DIY Tax is what a founder pays by becoming the AI implementer for their own business. The cash is small, a few thousand dollars, but it costs 60 to 120 hours of founder time, and far more once you price those hours honestly. The real bill is not the dollars. It is the founder turned into the bottleneck the rest of the business waits on.


In this article

Building AI into your business yourself looks like the responsible choice. The cash outlay is small, you keep control, and you learn how it all works on the way. Then the weeks start disappearing into reading docs, wiring tools together, and fixing whatever broke when you changed one setting. That time is the second hidden cost of AI, and almost no one puts it on the books. It is the second of seven taxes in the AI Tax, the real price of AI for a founder-led business once you count everything past the cash line. The first tax was the time you lose choosing tools. This one is the time you lose standing them up.

Should I set up AI tools myself or get help

Start with the honest number. Setting up AI yourself runs about $1,500 to $3,000 in cash plus 60 to 120 hours of the founder’s own time. The cash is the part you see. The hours are the part that costs you, and they come out of the scarcest resource in the business, which is the time of the person running it.

So the real decision is not “spend a little or spend a lot.” It is “whose time does this take.” When you do it yourself, it takes yours, the same time that should be going into customers, the team, and the work that grows the top line. That is the trade most DIY guides never price, because the cash looks so reasonable that the hours feel free. They are not free. They are the most expensive hours in the company.

Setting up AI yourself runs about $1,500 to $3,000 in cash plus 60 to 120 hours of the founder’s own time.
The Forest View, 2026

There is a fair counter-argument, and it is worth conceding. Doing it yourself teaches you how AI actually works, and that knowledge is genuinely useful. The catch is that learning is a one-time gain. The DIY Tax is the recurring cost of being the one who builds it, fixes it, and rebuilds it every time a tool changes underneath you. You can keep the lesson and still hand off the labor.

Why does setting up AI take so long

Because the slow part is not the AI, it is the integration. Connecting AI to the data and tools you already run, keeping the output consistent, and building a workflow the team will actually use needs skills most growing businesses do not have sitting around internally. When that skill is not on staff, the founder absorbs it, and the founder is learning it in real time, which is exactly why an afternoon turns into a week.

This is the part that quietly breaks the timeline. Buying a tool is fast. Making it work with the fifteen other things your business depends on is not, and it is the part nobody estimates honestly up front. The biggest delays come from data quality and readiness, not the technology, which is exactly the work that lands on the founder. We go deeper on this specific cost in the Integration Tax, because connecting the pieces is its own tax on top of building them.

Priced at the founder’s real rate, the slowness is not abstract. At $100 an hour, those 60 to 120 hours make the true cost roughly $6,000 to $12,000. State the assumption plainly: that figure assumes your time is worth $100 an hour, and most founders running a business would say theirs is worth more. Either way, the cheap option stopped being cheap somewhere around hour forty.

What does it really take to implement AI in a business

It takes more than the build. It takes the maintenance, and the maintenance is where DIY actually fails, because the time spent debugging and keeping it running can quietly exceed the time the AI saved you. The failure is rarely a dramatic crash. It is the ordinary afternoon you lose hours to a technical issue a specialist would have cleared in minutes, while the rest of the business waits on you to finish. That is the founder-as-bottleneck, and it is the real shape of the DIY Tax once the initial build is done.

The maintenance is where DIY actually fails: the time spent keeping it running can quietly exceed the time the AI saved.
DEV Community (AWS), The Hidden Cost of AI Coding, 2026

Think of it in three layers. There are the tools at the bottom, the work they do in the middle, and the business results at the top. Standing up AI yourself keeps you stuck at the bottom layer, building and fixing the plumbing, while the results at the top wait on you to climb back out. The most expensive version of this is the one where the founder never climbs out, because every change to a tool pulls them back down to rebuild.

That is the difference between learning AI and being trapped as its implementer. The first is a phase. The second is a job you accidentally gave yourself, and it is the one that taxes everything else you are supposed to be doing.

What a system that builds it for you looks like

The way out is not a better tutorial or a faster afternoon. It is to stop being the person who has to stand it up at all. Any real answer has to clear a clear bar: it has to build the workflow against your business rather than from a blank page, run the work across the tools you already use, and absorb the upkeep so a tool change does not pull you back down to rebuild. Your question changes from “how do I implement this” to “what do I want done.”

JynAI built Works, an AI Business OS, to clear exactly that bar. Here is the fit, plainly.

  • Pain: the build starts from a blank page and your own learning curve.
    Business-Aware Setup reads your LinkedIn, site, and files into a workspace that arrives already understanding the business, so there is nothing to stand up from scratch.
    Gain: the weeks you would have spent wiring it never get spent.

  • Pain: the work stalls because nobody on staff can connect the tools.
    Work That Actually Ships runs in three modes, Strategy to plan, Action to execute across the tools you already use, Automation to run hands-free, so the workflow finishes instead of waiting on you to build the next step.
    Gain: the workflow closes on its own instead of parking on your desk.

  • Pain: every tool change pulls you back down to rebuild.
    The Workflow Library and the model pool update underneath you, so when a new model ships it joins the pool and your existing work starts using it without you touching a thing, and your setup does not move.
    Gain: the rebuild that used to land on you every few months stops being your job.

  • Pain: you do not have a specialist on staff to keep it running.
    The Work Runs at the autonomy level you set, from approve-every-step to hands-free, so the upkeep is the system’s, not an afternoon stolen from your week.
    Gain: the founder stops being the thing the business waits on.

The affordability is the part that makes this honest rather than aspirational. The full capability set is available at the $49 tier, not behind an enterprise contract, so the founder who would have spent 60 to 120 hours building can reach the same ground without becoming the implementer to afford it.

And we are not theorizing. We were the implementers first. Machintel spent two years standing AI up and re-standing it up, the same hours that should have gone into running the business, before we built the layer that took the building off our plate. Once it was off, six teams were running on it inside ninety days. We are biased about our own product, of course. The argument under it does not need us: if the expensive part of DIY was always the founder’s weeks and not the cash, then a faster afternoon was never going to be the answer. Taking the building off the founder’s plate is.

The cash price of building AI yourself is the cheap part. Your weeks are the bill. The next time DIY looks like the responsible choice, price the hours before you price the tools, and ask whether the thing you are about to build is your business or just the scaffolding you will keep rebuilding around it.

Stop building yourself. Sign up for early access. Or size your own bill first with the AI Tax Calculator.

Common Questions

Should I set up AI tools myself or get help?

It depends on whose time you are willing to spend, which is the part most DIY guides never price. The cash to do it yourself is small, about $1,500 to $3,000, but it costs 60 to 120 hours of the founder’s own time, the scarcest resource in the business. If you do it yourself, keep the learning and plan to hand off the labor, because the labor is the part that recurs.

Why does setting up AI take so long?

Because the slow part is not the AI, it is the integration. Connecting AI to the tools and data you already run, keeping the output consistent, and building a workflow the team will use takes skills most growing businesses do not have on staff, and the biggest delays come from data readiness rather than the technology. The founder absorbs that work and learns it in real time, which is why an afternoon turns into a week.

What does it really take to implement AI in a business?

More than the build. It takes the maintenance, and the maintenance is where DIY quietly fails, because the time spent debugging and keeping it running can exceed the time the AI saved. The real cost is not a crash, it is the ordinary afternoon you lose to a problem a specialist would clear in minutes, while the business waits on you.

How much does it actually cost to build AI yourself?

The cash is roughly $1,500 to $3,000. The honest number adds 60 to 120 hours of founder time, which at $100 an hour puts the true cost closer to $6,000 to $12,000, and most founders would say their hour is worth more than that. The cash was never the expensive part.

What is the DIY Tax?

It is the second of the seven taxes in the AI Tax, the real price of AI for a founder-led business once you count past the cash line. The DIY Tax is what a founder pays by becoming the AI implementer: the hours to build it, the hours to fix it, and the founder turned into the bottleneck the rest of the business waits on.