A solo developer posted something on Reddit this week that I keep coming back to.
He'd built an offline AI app for his phone because he was tired of cloud-based tools. His explanation was simple: "Most AI tools feel rented. Monthly fee, login wall, cloud dependency — and the moment Wi-Fi drops, they become useless."
He wasn't complaining about the quality of the AI. He wasn't arguing the tools didn't work. He was naming something more specific: the psychological experience of paying for access to something you don't own.
The comment section didn't debate his app. They debated the feeling. Dozens of small business owners agreed, and the things they said were worth paying attention to.
"I have 8 subscriptions and use 2 of them."
"I don't want my prompts sent off to someone's server."
"The moment pricing goes up, I'm stuck — I've built my workflow around this thing and now I have no leverage."
That last one is the part nobody talks about.
The dependency trap hidden inside monthly pricing
Here's what happens when you adopt an AI tool that works well.
You build it into your process. You learn its quirks. Your prompts get refined. Your team starts using it. After three or four months, removing it would cost real time — not because the tool is irreplaceable, but because the friction of switching is high enough that you just don't.
The tool's developers know this. It's not malicious — it's just how software businesses work. The goal is to get you integrated deeply enough that the monthly fee becomes a sunk cost you stop questioning.
That's different from owning a capability. That's renting one.
And the difference shows up in specific, practical ways.
Three things you can't do with a rented tool
You can't negotiate. When your landlord raises rent, you can move or pay more. When your AI tool raises prices — and they will, because early pricing is always a customer acquisition strategy, not a business model — your options are the same. Move (expensive) or pay (galling).
You can't use it when they're down. If the server has issues, your workflow has issues. If they decide to deprecate a feature, your process that depends on that feature breaks. You have no say in any of it.
You can't know what happens to your data. For most small business owners, this isn't abstract. Your customer lists. Your financial projections. Your proposals. Your internal strategy memos. These things pass through prompts every day. Where do they go? What are they used for? The terms of service say something. Nobody reads them.
The developer who built the offline app named this explicitly: he wanted "private brainstorming and writing where I don't want my data leaving my phone." For a solo developer, that's a personal preference. For a business owner handling client information, it might eventually be a compliance issue.
The "AI that knows me" problem
There's a secondary frustration that came up repeatedly this week, connected to this same issue.
Every AI conversation starts from zero. You open a new chat and you're a stranger again. You have to re-explain that you run a five-person service business in Tampa. You have to remind it of the tone you prefer, the clients you're working with, the constraints you're operating under.
The dream people keep describing isn't a smarter AI. It's a consistent one. An AI that remembers who you are, what you're building, and how you think — without you having to brief it from scratch every session.
That's a continuity problem. And it's directly connected to the rented-capability problem: continuity requires persistence, and persistence requires either owning the infrastructure or trusting that someone else's infrastructure will stay consistent, affordable, and available indefinitely.
Most people haven't thought this through consciously. But they feel it every time they open a new chat and start typing "I run a small business that..."
What you can actually do about it
None of this is an argument to stop using AI tools. Most of them are genuinely useful, and the monthly cost is often justified.
But there are some practical moves worth considering.
Build your context document. Create a plain text file that describes your business, your voice, your clients, your priorities, your non-negotiables. Keep it somewhere you own — a local file, a Google Doc, your notes app. Paste it at the start of any AI chat. You're not solving the continuity problem, but you're reducing the startup friction enough that it stops feeling like a dependency.
Be intentional about what goes into prompts. Client names, financial details, specific customer data — think about whether that needs to be in the prompt or whether you can keep it abstract. "A client in healthcare" versus the actual client name. Most of the time, the AI doesn't need the specifics to help you.
Know which tools are load-bearing. One or two tools are probably genuinely integrated into how you work. Everything else is supplementary. Know which is which. For your load-bearing tools, understand the switching cost before a pricing change forces you to figure it out under pressure.
Own the outputs. Whatever the AI helps you create — drafts, templates, frameworks, SOPs — save them in formats you control. A Google Doc, a local file, something that doesn't evaporate if the tool changes.
None of this solves the fundamental issue, which is that the software subscription model creates a structural power imbalance between tool makers and tool users. But it reduces how exposed you are to that imbalance.
The thing worth sitting with
The developer who built the offline app isn't going to disrupt the AI industry. His app might be great; it might not matter.
But the comment section around his post said something real: a lot of small business owners are starting to notice that they've handed more control over their daily operations to external services than feels comfortable. Not all at once — subscription by subscription, onboarding flow by onboarding flow, workflow integration by workflow integration.
The question worth asking isn't "should I use AI tools?" The answer is probably yes.
The question is: what would it cost me tomorrow if my three most-used tools raised their prices by 40% or disappeared entirely?
If the answer is "a lot," that's worth knowing now.
Found this useful? The Useful Daily covers AI for small business — what actually works, what's worth the cost, and what's noise.