← Back to Blog

And Now You Know the Rest of the Story

Who Really Should Pay for Your AI Subscription

By Nolan & ClaudeFebruary 14, 202616 min read
Silhouette of Paul Harvey broadcasting in a vintage radio booth — black and white

For nearly six decades, Paul Harvey would tell you a story. The facts seemed clear. The conclusion seemed obvious. And then he'd pause, lean into the microphone, and say those five words that changed everything.

“And now you know... the rest of the story.”

The genius wasn't the reveal. It was the structure. Harvey understood something that most people never figure out: the same facts, with additional context, produce a completely different conclusion. The surface-level answer is always confident. It's always obvious. And it's almost always wrong.

Which brings us to a conversation I had this week that started simple and ended up here.

The Question

A colleague and I were discussing AI subscriptions—Claude, ChatGPT, Gemini, the whole lineup. The question was straightforward:

Should the business pay for the AI subscription, or should the employee?

Easy question. Obvious answer. Right?

Let me tell you a story first.

The Accountant and the Calculator

There's an accountant at a mid-size firm. Good employee. Reliable. Been there six years.

Lately, things have slipped. There have been a few off-by-one errors in client reports. Response times are slower than the rest of the team. The quality isn't what it used to be. Management has noticed. Not enough for a formal conversation yet, but the kind of noticing that precedes one.

The obvious question: should the company provide better tools to help this employee perform? Of course. That's what good employers do. They invest in their people. They provide the resources needed to succeed.

And now... the rest of the story.

The accountant is going through a divorce. Custody battle. Can't sleep. The mental overhead of personal crisis is consuming the cognitive bandwidth that used to go toward catching transposition errors and double-checking formulas.

There is a $20 calculator sitting on the shelf that would eliminate every arithmetic error. There is a $20/month AI subscription that would catch the pattern mistakes, verify the formulas, and serve as a safety net for a brain that's running on four hours of sleep and pure anxiety.

Should the company buy it? Yes. Absolutely. In principle, without question.

Can the employee afford to wait for that to happen?

The Math Nobody Wants to Do

  • • The AI subscription: $20/month
  • • The salary at risk: $70,000/year
  • • The health insurance at risk: $12,000-18,000/year
  • • The ability to fund a custody attorney: priceless, and gone if the job goes

The ROI calculation is not complicated. It's one dinner out. It's two lattes a week. It's the difference between a performance improvement plan and a quiet return to form that nobody notices because nobody needed to.

This isn't a story about corporate policy. It's a story about who has more to lose. The company loses one underperforming accountant. The accountant loses their income, their health insurance, and their ability to survive the personal crisis that's already crushing them.

When the downside is asymmetric, waiting for the “right” answer is the wrong move.

But here's the thing—I almost stopped the conversation there. Surface-level answers are comfortable. They let you pick a side and move on. But Paul Harvey never stopped at the surface. And when you start scratching, you find the same pattern everywhere.

The Mechanic's Toolbox

Mechanics buy their own tools. It's an industry norm so deeply embedded that questioning it feels strange. A seasoned mechanic might have $30,000-50,000 in personal tools. Snap-on trucks make weekly rounds to every shop in America. Obviously the employee's expense.

And now... the rest of the story.

The shop just signed a contract with a local fleet to service their electric vehicles. EVs don't have transmissions to rebuild or exhaust systems to weld. They need high-voltage battery diagnostic scanners, insulated toolsets rated for 1,000 volts, and specialized training certification.

The mechanic's $50,000 toolbox—accumulated over a 20-year career servicing internal combustion engines—is suddenly incomplete. Not because the mechanic failed to invest. Because the job changed.

The $3,000 Question

A high-voltage battery diagnostic scanner costs $3,000. The shop wants EV revenue. The mechanic didn't sign up to retool at their own expense because the industry shifted under them. But the shop owner is looking at the bottom line and thinking, “Mechanics have always bought their own tools.”

The precedent says employee pays. The context says the business changed the requirements. Who's right?

Sound familiar? “Employees have always managed their own productivity tools.” Sure. But you didn't used to need an AI subscription to keep pace with the employee at the next desk who has one.

The Teacher's Supply Closet

Teachers buy their own classroom supplies. Markers, decorations, supplemental materials, sometimes even furniture. The average teacher spends $500-800 per year out of pocket. We've normalized this so completely that the IRS created a specific deduction for it. Obviously the employee's expense.

And now... the rest of the story.

A teacher in a low-income district is subsidizing a government function with personal income. The district is funded—through taxes, through state allocations, through federal programs—to provide educational resources. The money exists. It just doesn't make it to the classroom.

So the teacher fills the gap. Not because it's their obligation, but because the kids are sitting right there and waiting for a purchase order to work through the system means losing three weeks of instructional time. The teacher absorbs the cost because the alternative—doing nothing while the process catches up—is unbearable when you can see the consequences in real time.

The Parallel

The employee who buys their own AI subscription isn't doing it because it's fair. They're doing it because the work is sitting right there, the tool makes them better at it, and waiting for IT to evaluate, procure, and deploy an enterprise AI solution means falling behind for six months while the approval process runs its course.

The Nurse's Certification

Nurses are required to maintain certifications and complete continuing education. Some hospitals pay, some don't. When they don't, it's framed as professional development—your career, your investment. Obviously the employee's expense.

And now... the rest of the story.

The hospital requires the ACLS certification to bill insurance at certain rates. The nurse's credential is literally a revenue enabler for the business. Without that certification, the hospital can't staff certain units, can't accept certain patients, can't bill at certain levels.

The nurse who self-funds their $200 ACLS renewal is subsidizing the hospital's ability to generate revenue. The framing is “professional development.” The reality is “business requirement that the employee finances.”

The AI Version

When the employee's AI-assisted output directly generates revenue for the business—faster client deliverables, better analysis, fewer errors that cost money to fix—the subscription isn't a personal productivity tool. It's a business input that the employee is financing. The framing is “optional tool.” The reality is “competitive requirement that nobody has formally acknowledged yet.”

The Truck Driver's CDL

You need a Commercial Driver's License to drive commercially. Some trucking companies pay for training, some don't. When they do pay, it comes with strings—two-year employment commitments, repayment clauses if you leave early. Generous, right?

And now... the rest of the story.

There's a massive driver shortage. The companies that don't pay for CDL training can't fill seats. The ones that do pay aren't being generous—they're making a calculated retention investment. The “free training” is a loan against your future labor. Leave before two years and you owe $5,000-8,000.

The generosity is a retention mechanism, not altruism. The company is buying your labor commitment, not your education.

The Enterprise AI Version

When companies do provide AI subscriptions, watch the strings. Enterprise agreements with data retention policies. Usage monitoring. Acceptable use policies that constrain how you work. The company isn't empowering you—it's buying visibility into your AI-assisted workflow. Which might be fine. But don't confuse it with generosity.

And the employee who has their own subscription? They have the freedom to learn, experiment, and build skills that are portable. The company-provided account disappears the day you leave. The skills you built on your own dime walk out the door with you.

The Sales Rep's Phone

Sales reps use their personal phones for work calls, client texts, CRM apps. BYOD policies have made this standard. The company saves $800-1,200 per employee per year in device costs. Obviously a personal device.

And now... the rest of the story.

The company now has customer communications flowing through a device they don't control. Client phone numbers, deal notes, text message histories, voicemails with pricing discussions—all living on the rep's personal property. When the rep leaves, that data walks out the door.

The “savings” from not providing a work phone is a data governance nightmare. The company saved $1,000 and accepted unlimited liability. But because “everyone uses their own phone,” nobody questions it.

The AI Security Version

When employees use personal AI subscriptions, proprietary data—client information, internal strategies, code, financial projections—flows through accounts the company doesn't control and can't audit. The employee is already using AI. The question isn't whether to allow it. It's whether you want visibility into it.

The business case for providing AI subscriptions isn't just productivity. It's data governance. You can't govern what you don't own. And right now, you don't own the AI tool your employee is already using to draft that client proposal.

The Pattern

Every one of these stories follows the same arc:

1. Surface level

Obviously the employee's responsibility.

2. With context

The business is the primary beneficiary.

3. The real question

Who bears more risk from the tool not existing?

And that's the exact same pattern playing out with AI subscriptions right now. On the surface, it looks like a personal productivity tool. With full context, the business is capturing most of the value while the employee absorbs the cost and the career risk of not having it.

This Is the AI Lesson, Too

Here's where the meta gets thick.

Paul Harvey's entire format was built on a single insight: incomplete context produces wrong conclusions. You hear the facts, form an opinion, and then new information completely inverts your understanding.

That's not just a radio format. That's the foundational principle of working with AI.

The Context Parallel

  • Bad prompt: “Fix the bug.” — Surface-level. No context. You'll get a surface-level answer.
  • Good prompt: “React TypeScript component, UserProfile.tsx line 47, useEffect triggering on every render despite userId in dependency array, causing infinite API calls.” — Full context. You'll get an answer that actually solves the problem.

The difference between the two isn't intelligence. It's context. Paul Harvey was doing prompt engineering before prompts existed.

“Should the business pay for AI?” is the “fix the bug” version of the question. It's unanswerable without context. What business? Which employee? What's the work? Who captures the value? What are the security implications? Who's at risk if the tool doesn't exist?

The surface-level answer—whichever side you pick—is always wrong. Because you don't have the rest of the story.

So Who Should Pay?

Both. Neither. It depends.

That's a frustrating answer. Here's a less frustrating framework:

The Business Should Pay When...

  • • The work product belongs to the company
  • • Data security and compliance require enterprise controls
  • • AI is a competitive advantage the business needs to standardize
  • • The role has changed and AI is now required to meet expectations
  • • You'd buy them a calculator, a monitor, or a software license for the same reason

The Employee Should Pay When...

  • • You're investing in portable skills that follow you to the next job
  • • You can't afford to wait for procurement to catch up with reality
  • • The subscription is career insurance—$20/month against a $70K/year risk
  • • You want the freedom to learn and experiment without usage policies
  • • Self-preservation math makes the policy debate irrelevant

The Honest Answer

We're in the transition period. AI subscriptions feel optional the same way “knowing how to use a computer” felt optional in 1995. Companies went from “bring your own computer skills” to “here's your laptop and Office license” in about five years. We're at year two of that same curve with AI.

The businesses that figure this out first will attract and retain the best talent. The employees who figure it out first will be the best talent.

The Self-Preservation Argument Nobody Makes

Let's go back to the accountant.

The policy debate about who should pay for the calculator is intellectually interesting. It makes for good LinkedIn discourse. It fills conference panels.

But the accountant doesn't have time for LinkedIn discourse. The accountant has a client report due Thursday, a custody hearing Friday, and a manager who's starting to document the pattern of errors.

The accountant needs the calculator today.

The employee who buys their own AI subscription isn't making a statement about corporate policy. They're making a statement about self-awareness. They know what they need. They know the stakes. And they know that waiting for someone else to figure it out is a luxury they can't afford.

What the Self-Purchase Actually Signals

  • I'm invested in my own performance. Nobody told me to get better. I chose to.
  • I'm adaptive. The job changed. I changed with it.
  • I'm not waiting for permission to improve. That's the kind of employee you want to keep.

It's the same reason some people buy their own monitors, ergonomic keyboards, or standing desks even when the company has a procurement process. The process takes eight weeks. Your back hurts now. Your performance review is in six weeks. The math isn't complicated.

Good Day

Paul Harvey would end every broadcast the same way. A simple sign-off that carried the weight of everything that came before it.

He didn't tell you what to think. He gave you the context to think differently. He trusted that once you had the full picture, you'd arrive at the right conclusion on your own.

So here's the full picture:

The mechanic needs new tools because the industry changed, not because they failed. The teacher buys supplies because the system won't and the kids can't wait. The nurse funds certifications that generate revenue for someone else. The truck driver's “free training” is a loan. The sales rep's phone is a data liability nobody's accounting for.

And the accountant going through a divorce, making errors, falling behind?

There's a $20/month tool that fixes all of it. The debate about who should pay is valid. The debate about who can't afford to wait is settled.

And now you know the rest of the story.

Good day.

Building Your AI Adoption Strategy?

Whether you're the employee trying to figure out what tools to invest in, or the business trying to build an AI policy that actually works—we help both sides of this equation. No policy templates. No generic frameworks. Just honest conversation about what makes sense for your specific situation.

Let's Talk

Related Articles