A while back I had a stupidly simple task: turn a few photos into one PDF. That’s it. I searched “images to PDF,” clicked the first “free” result, dragged my files in, and waited. Then came the watermark — a grey diagonal stamp across every page, with a cheerful little note offering to remove it for $9 a month.
So I closed the tab and tried the next one. That one didn’t watermark — it just wanted me to create an account and verify my email before I could download the file I’d already made. The third let me download exactly two files before a wall slid up: daily limit reached, upgrade to continue. A five-minute job took forty, and I never actually finished it. Somewhere in there I got annoyed enough to stop and ask the question I’d never bothered with: why is “free” never actually free?
It’s not bad luck — it’s the business model
Once I looked, the pattern was everywhere, and it was always one of the same few moves:
- The watermark. The conversion is free; a clean copy costs money.
- The daily limit. Two files free, then a paywall — conveniently right when you’re in a hurry.
- The signup wall. Free to convert, but you hand over an email to download.
- The feature lock. The basic format is free; the one you actually came for is “Pro.”
- The size cap. Fine for a two-page memo, useless for the 40-page report you actually have.
None of it is a scam, exactly. But “free” with that many asterisks isn’t really free — it’s a free sample.
Then I figured out why — and it bothered me more
Here’s the part that changed how I think about these tools: almost all of them do the work on their own servers. You upload your file, their machines process it, you download the result. Servers cost real money to run — so the tool has to earn that money back, and the gates above are how. The watermark and the rent are the same thing.
But that means there’s a second catch, the one you never see: your file leaves your computer. To convert that PDF, you handed it to a company’s servers. For holiday photos, who cares. But I started thinking about the other things people run through these tools every day — a signed contract, a tax return, a medical form, a photo with their home address baked into the GPS data — all uploaded to a stranger’s machine just to change a file format. That’s a lot of trust for a five-minute task.
So I built the thing I wanted
I’m one person, not a company. But I could build, so I built the version I’d been looking for: file tools that run entirely in your browser. Your file is processed on your own device and never gets uploaded anywhere.
What surprised me is that going client-side fixed both problems at once. No servers to pay for meant there was nothing to gate — no watermark, no daily limit, no signup, because there’s no cost for me to recover. The “free” and the “private” turned out to be the same design decision. That became AntiUpload: around 50 tools now — convert, compress, merge, OCR, sign — all free, all in your browser, none of them touching your files.
How to check any tool yourself
You shouldn’t take my word for this — not even about my own tool. Three quick ways to tell:
- Look for the asterisks first. If the homepage is already pitching “Pro,” you know the play before you upload anything.
- Open the Network tab (press F12, then Network) and run a file through it. If your file is being uploaded, you’ll see it leave. A genuinely local tool won’t send it anywhere.
- Go offline. Load the page, kill your internet, and try it. If it still works, the work is happening on your machine — not on someone’s server.
The honest bottom line
I’m not anti-paying. Adobe and the serious players earn their subscriptions on genuinely advanced features, and real professional work is worth paying for. But the everyday stuff — convert this, compress that, merge these — shouldn’t cost you a subscription or your privacy. That should just be free, and it should stay on your device.
I built one version of that. Even if you never use it, now you know what “free” is usually hiding — and that it doesn’t have to be.
The same question, one more place
File tools aren’t the only thing that quietly asks for your documents. The moment you paste a sensitive file into an AI chatbot, the same question applies: where does it actually go? I wrote about how to think that through in Is it safe to put confidential documents into AI chatbots?