At some point in the last few years, you’ve probably scanned a document with your phone — a passport for a visa application, a W-2 for taxes, a lease agreement, a medical form. The app made it fast: point the camera, the edges auto-crop, the text gets recognized, and you have a clean PDF in seconds.
What most people don’t think about is where that document went in order to become a clean PDF — and where it stayed afterward.
How Document Scanning Apps Actually Work
The “magic” of a scanning app — auto-crop, perspective correction, glare removal, and especially text recognition (OCR) — requires processing. That processing happens in one of two places: on your device, or on the app provider’s servers.
On-device processing means the image never leaves your phone to be cropped, enhanced, or read. The app’s OCR model runs locally, using your phone’s processor, and the resulting file is created and stored without any of that data touching the network — unless you choose to share or back it up afterward.
Cloud processing means your scan — the actual image of your passport, tax form, or contract — is uploaded to the provider’s servers, processed there (often using more powerful OCR models than a phone could run), and the result is sent back to your device. The original image, the extracted text, or both, may be retained on those servers afterward, governed by the provider’s retention policy.
The distinction matters enormously for sensitive documents, and it’s usually not obvious from using the app which one is happening.
What Popular Scanning Apps Actually Do
CamScanner, one of the most widely used scanning apps globally, processes documents on the company’s cloud servers. This is how it achieves strong OCR accuracy across many languages and document types — but it also means every document scanned through the app, including IDs, passports, and financial paperwork, is transmitted to and processed by CamScanner’s infrastructure.
CamScanner has a notable history worth knowing about: in 2019, the app was temporarily removed from the Google Play Store after security researchers discovered a malicious advertising SDK bundled with it, capable of downloading and executing additional code on users’ devices without consent. The malicious component was removed and the app was reinstated, and there’s no indication of similar issues since. But the incident is a useful reminder that an app processing your most sensitive documents in the cloud is also an app whose entire security posture — including third-party code it bundles — becomes part of your exposure.
Adobe Scan takes a different approach: scans sync directly to Adobe Document Cloud, tied to your Adobe account, making them accessible across devices and within Acrobat. This is convenient if you’re already in the Adobe ecosystem, but it means your scanned documents become part of your Adobe Document Cloud storage, governed by Adobe’s broader account and data policies — the same account that may be linked to other Adobe products and services.
By contrast, apps built around local-first processing — Genius Scan and similar tools — perform OCR and document creation entirely on-device, with documents never leaving your phone unless you explicitly export or share them.
None of this makes cloud-based scanning apps unsafe for everyday use. It means the privacy-relevant question — “where does this image go, and what happens to it there?” — has a different answer depending on which app you’re using, and that answer rarely shows up anywhere in the scanning interface itself.
Not All Scans Are Equally Sensitive
It’s worth being deliberate about which documents matter most here, because the answer isn’t “treat every scan with maximum paranoia” — it’s “match the handling to the sensitivity.”
Low sensitivity: A receipt for an expense report. A flyer you want to remember. A whiteboard photo from a meeting. If these end up processed in the cloud, the consequences of exposure are minor.
Medium sensitivity: Signed contracts, utility bills, insurance documents. These contain personal information — your name, address, account numbers — but aren’t typically targets for identity theft on their own.
High sensitivity: Passports, driver’s licenses, Social Security cards, tax returns, bank statements, medical records. These documents, individually or combined, contain everything needed for identity theft, and several of them (passport and government ID scans especially) are exactly the kind of data that’s valuable on the black market and frequently targeted in breaches.
The practical implication: for the high-sensitivity category, the choice of scanning method and storage location matters more than for the rest. A tax return scanned and processed through a cloud OCR pipeline, then stored indefinitely on that provider’s servers, has traveled through and rested in more places than the same tax return scanned on-device and moved directly into storage you control.
What Happens to a Scan After OCR Runs
This is the question that’s hardest to answer and most worth asking. Once a cloud scanning service has processed your document and returned the result to you, what happens to its copy?
The honest answer is: it depends on the provider’s retention policy, and that policy is usually not specific to scanned documents — it’s the same general data retention policy that governs everything else the company stores. “We retain data as long as necessary to provide the service, or as required by law” is the typical language. It doesn’t tell you whether your passport scan from eight months ago still exists on a server somewhere.
For documents you scan once and need to keep — tax records you might need for years, ID documents you reference occasionally — this matters more than for documents you scan, use immediately, and never need again. A scan that lives only as long as the immediate task is a different risk profile than a scan that’s quietly retained indefinitely as a side effect of having used a cloud OCR pipeline once.
A Practical Workflow for Sensitive Documents
You don’t need to give up the convenience of scanning apps to handle sensitive documents more carefully. A reasonable approach:
For high-sensitivity documents (IDs, passports, tax forms, financial records): Use an app with on-device OCR processing where possible, or accept that cloud OCR is being used and plan to delete the scan from that provider’s storage afterward — don’t assume it’s automatically removed. Then move the final file to storage you control and trust for long-term keeping.
For medium-sensitivity documents: Cloud-based scanning apps are reasonable for convenience. If you need to keep the document long-term, the same principle applies — move it to your primary storage rather than leaving multiple copies scattered across whichever app happened to scan it.
For low-sensitivity documents: Use whatever’s most convenient. The stakes are genuinely low.
The common thread is treating the scanning app as a temporary tool — the thing that converts a physical document into a digital one — rather than as the permanent home for that document. Where the file ends up living long-term is a separate decision, and it’s the one that matters most for anything sensitive.
Where the Finished Scan Should Live
Once you have a clean PDF or image of an important document, where it’s stored long-term determines who can access it, how it’s protected, and what happens to it over time. The same questions that apply to photo storage apply here, often with higher stakes given the contents:
- Is the file encrypted in transit and at rest?
- Does the storage provider’s business model create any incentive to analyze document contents?
- Can you organize and retrieve the document easily when you actually need it — at tax time, during a loan application, when renewing a passport?
- Is there a clear policy on what happens if you delete it?
A scanned passport sitting in a generic “Downloads” folder, synced automatically to whatever cloud service your phone defaults to, has often ended up somewhere you didn’t choose and wouldn’t have chosen if asked directly.
How daftei Fits Into This Workflow
daftei is designed as the destination for documents like this — not the scanning tool itself, but the place scanned files go once they exist. Documents uploaded to daftei are encrypted in transit with TLS 1.3 and at rest with AES-256, organized alongside your other personal files and memories rather than scattered across whatever app last touched them.
daftei doesn’t run OCR or cloud document analysis on your files — it stores what you give it. There’s no advertising business model creating an incentive to look inside your tax records, and daftei never trains AI models on user content. The 5 GB free tier is enough for years of important documents for most people; Pro removes the limit entirely at $5.99/month or $44.99/year.
The workflow that makes sense: scan with whatever tool is convenient for the task, then move anything you need to keep — especially anything in the high-sensitivity category — into storage that’s organized, encrypted, and under your control, rather than leaving it as a byproduct in the scanning app’s own storage.
The Five-Minute Audit
If you’ve been scanning documents with your phone for a while, it’s worth a five-minute check: open your scanning app, look at how many documents are stored there, and ask whether you know what kind of processing each one went through and where the originals might still exist.
For most people, the answer is “I have no idea” — which isn’t a crisis, but is exactly the kind of gap worth closing for the documents that would actually cause a problem if they were exposed. Moving your passport scan, your tax PDFs, and your signed contracts to a single, deliberate, encrypted storage location takes a few minutes and closes that gap for good.