If you've opened your security camera app this year and felt like the free version does less and the paid version costs more, you're not imagining it — and it isn't one company being greedy. It's the same economic squeeze hitting every subscription camera vendor at once. Here's the honest, vendor-by-vendor scan of what actually changed in 2026, why it's happening to all of them simultaneously, and the one option most people don't realize they already own.
What actually changed, vendor by vendor
The headlines are easy to dismiss as routine price creep. Lined up together, they tell a clearer story: across the major vendors, 2026 has been the year the free tier shrank and the paid tier climbed — at the same time.
- Wyze — long the budget darling — raised Cam Plus Annual in March 2026 from $19.99 to $29.99 per camera per year, and now steers multi-camera households toward Cam Unlimited at about $9.99/month (roughly $99/year). The useful stuff — longer cloud event history, the better person/package AI events — increasingly lives behind that line, and the no-subscription experience keeps losing features it used to include.
- Arlo — the basic Arlo Secure single-camera plan moved from $4.99 to $7.99/month (about $96/year), as widely reported. Single-camera plans have been quietly de-emphasized in favor of the pricier multi-camera tiers, and core conveniences are gated behind the subscription.
- AlfredCamera — tightened its free tier into something closer to a trial: a small cap on the number of cameras, shorter clip retention, time-limited live sessions, and watermarked exports. The paid plan's annual price also rose (the widely reported jump to around $35.99/year). The free tier is now essentially a guided tour of the paywall.
- Eufy — marketed for years as the "buy the hardware, no monthly fee" option — has steadily layered cloud storage and AI-feature charges on top. Facial recognition, extended event history, and richer notifications increasingly sit behind an optional cloud plan, eroding the no-subscription pitch that made the brand popular.
Four different companies, four different product philosophies, one identical direction of travel. That's not a coincidence, and it's not a coordinated cartel either. It's arithmetic.
Why it's happening to all of them at once
Here's the mechanism, because once you see it the pattern stops feeling like bad luck and starts feeling inevitable.
A cloud-backed camera is genuinely expensive to run. Every frame your camera uploads is bandwidth the vendor pays for. Every clip it retains is storage the vendor pays for. Every "watch from anywhere" session and every ML motion-alert is compute the vendor pays for. For a popular app with hundreds of thousands of cameras online, that's a five-, six-, or seven-figure bill every single month — and it arrives whether or not the user ever pays a cent.
There are only four ways to cover a recurring bill like that:
- Charge a subscription. The honest option, and the one the reputable vendors take.
- Run ads — which on mobile usually means embedding leaky third-party SDKs that harvest data.
- Sell your data or metadata to make the "free" tier pay for itself.
- Shut down — which stranded-camera users discover the hard way.
When the monthly cloud bill grows faster than free users convert to paid, a vendor has exactly two levers left: raise prices or shrink the free tier. In 2026 they've all been pulling both. It isn't malice. It's the unavoidable consequence of building a product whose core feature — "your video, on our servers, watchable from anywhere" — is also its biggest permanent cost.
Which means the price hikes aren't a bug you can wait out. They're the business model working as designed. Next year's version of this article will have different numbers and the same shape.
The structural risk hiding inside "watch from anywhere"
There's a second cost to the cloud-relay model that doesn't show up on the invoice, and 2026 made it concrete.
On May 11, a firmware exposure in Meari-based devices left roughly 1.1 million cameras across 378 brands watchable by anyone who extracted a single hardcoded key. One key, more than a million living rooms. That is the structural risk of the cloud-relay design: if your video lives on a vendor's server so you can watch it from anywhere, then a breach of that server is a breach of your home. You did nothing wrong, your password was fine, and your nursery feed was still exposed — because the architecture put it somewhere a stranger could reach.
Every "camera footage leaked" headline you've ever read had a cloud account in the middle. The convenience and the vulnerability are the same feature viewed from two sides.
The option most people already own
If you're done paying — and done being one breach away from exposure — there's an entire category of camera app that has none of this pressure, because it has no cloud bill in the first place.
These apps record to the phone's own local storage and let you view the feed over your home Wi-Fi through a small built-in web server. No vendor server means no monthly bill means nothing to raise or shrink — and no central server to breach. The hardware is something almost everyone already has: that old Android phone in a drawer. A Pixel 3a, an aging Galaxy, a hand-me-down on Android 9 or newer. It has a good camera, a battery, Wi-Fi, and a processor that once ran a whole smartphone. As a security camera it's wildly overqualified, and it costs nothing because you already own it.
The honest trade-offs, stated plainly so nobody feels misled:
- No view-from-anywhere out of the box. By default you see the camera when you're on your home network. If you genuinely need to peek in from outside, you set up a free VPN back into your own home or route through your own YouTube Live channel — more work, still zero recurring dollars, and the footage never touches a third party.
- No vendor-server ML motion-alerts, because there's no vendor server running the model.
If those two features are worth ~$100/year to you, a subscription is the honest way to buy them — and I'll say that plainly rather than pretend otherwise. But if what you actually want is "point a camera at the back door, the crib, the driveway, or the garage and watch it on my phone at home, for free, without anyone holding my video on a server," an old Android phone plus a local-only app does exactly that.
The app I work on, Background Camera RemoteStream, is built for exactly this pattern: it's free (the whole product is free; a one-time Pro purchase only adds optional YouTube Live for off-LAN viewing), records with the screen off, keeps everything on the device, serves the LAN feed behind a PIN, and asks for no Location, no Contacts, and no account. You can grab it on Google Play. IP Webcam and Haven are other apps in the same no-cloud category — I mention them because the point here is the category, not just my app.
How to verify "free" actually means free — in 60 seconds
"Free, no cloud" is easy to print and hard to trust, so don't trust it — check it. Open Settings → Network / Data usage on the camera phone, find the app, and watch its background data while it records. On a true local-only app that number sits near zero, because nothing is being uploaded. If a so-called "free" app is quietly streaming your footage to a server, the data counter is where it shows up.
That single check is the whole difference between "free because it's efficient" and "free because you're the product." A camera that uploads is a camera that costs someone money to run — which is exactly why, sooner or later, it costs you money too, in dollars or in data. The full architectural fork behind the word "free," plus the longer version of this self-audit, is here: Is It Possible to Use a Free Android Camera App Without Giving Up Your Privacy? The Architectural Fork Behind "Free".
The honest cost comparison
Here's the per-year picture for a single camera, recurring costs included:
| Setup | Year-1 cost | Recurring | Notes |
|---|---|---|---|
| Old Android phone + free local-only app | ~$0–$3 (electricity) | none | Reuses hardware you own; recordings local; LAN viewing behind a PIN; no cloud server to breach |
| Wyze Cam + Cam Plus Annual | camera + $29.99/yr | yes | Annual raised from $19.99 in March 2026; multi-cam steered to ~$99/yr Unlimited |
| Arlo + Arlo Secure | camera + ~$96/yr | yes | Reported $7.99/mo per-camera plan; cloud relay |
| AlfredCamera (free → paid) | $0 → ~$35.99/yr | yes | 2026 free tier capped, watermarked, time-limited; paid plan raised |
| Eufy + cloud/AI features | camera + growing fees | growing | "No-fee" positioning eroded by add-on charges |
The old-phone row isn't a rounding trick. The only ongoing cost of a phone left on a charger is the electricity it draws — call it a couple of dollars a year. Everything that makes the other rows expensive is the cloud subscription, and a local-only setup simply doesn't have one. If you want the cheapest-possible build spelled out step by step, I walked through the whole $0 setup here: What's the Cheapest Way to Set Up a Home Security Camera Without a Subscription in 2026?.
The catch nobody mentions: keeping it alive
I'd be doing you a disservice if I stopped at "install an app, you're done." There's one genuinely technical hurdle, and it's why a lot of DIY "old phone camera" experiments fizzle: modern Android aggressively kills background work to save battery. A naïvely built camera app can record beautifully for three or four hours and then silently die — and you find out at the exact moment you needed the footage.
Beating that requires the app to own a proper foreground service, handle wake locks correctly, and configure its Camera2 session to survive the OS's power management. A free app doesn't have to get that right; a good one does. If you want the full breakdown of why old-phone cameras die after a few hours and how a well-built one fixes it, it's here: Why Your "Old Phone Security Camera" Dies After 4 Hours (And How to Fix It on Modern Android). Read it before you rely on any DIY camera — it's the difference between a toy and something you'd trust on your front door.
So — is it worth switching?
If your plan got worse this year, that wasn't an accident you can wait out; it's the cloud-camera business model doing exactly what it's built to do. The price went up because the vendor's monthly bill went up, and that bill never goes away.
The escape is the same regardless of which logo is on your current camera: a phone from your drawer, a free local-only app you can audit in 60 seconds, and a charger. No recurring invoice, and — just as importantly in a year that put 1.1 million cameras one key away from exposure — no vendor server holding your footage for someone to breach.
You don't have to fire your subscription today to find out. Dig out the old phone, install a local-only app, and run the data-usage test for yourself. The app is free on Google Play, and there's more on the project at superfunicular.com.
Background Camera RemoteStream is a free, local-only Android camera app by Super Funicular LLC — record with the screen off, view over your own Wi-Fi in any browser behind a PIN, no account and no cloud. Built in the open with Claude Code over 75+ AI-assisted sessions.
Top comments (0)