20 privacy concepts everyone who uses a smartphone should know

From metadata to zero-day exploits, these 20 terms explain how your phone collects, shares, and exposes your data — and what you can do about it

Your smartphone knows more about you than almost any other object you own. It tracks where you sleep, who you call, what you search for, what you buy, and how you move through the world. Most of that information flows somewhere — to app developers, advertisers, data brokers, and occasionally to governments or hackers. Yet the vocabulary used to describe these processes is often technical, buried in privacy policies, or deliberately obscured by the companies that benefit from your data.

Privacy as a concept has shifted dramatically over the past two decades. In the early days of mobile computing, concerns centered on whether someone could read your text messages. Today the surface area of exposure is vastly larger. A single smartphone app can request access to your location, contacts, microphone, camera, clipboard, and calendar — often without a clear explanation of why it needs any of those things. The permissions system that was designed to protect you has, in practice, become a mechanism through which companies extract enormous amounts of behavioral data with your nominal consent.

The companies that build operating systems, apps, and advertising networks have developed sophisticated technical systems for collecting and monetizing that data. Many of these systems operate invisibly, in the background, without any obvious sign that they are running. Understanding them requires knowing the terms — not because the terms are inherently useful, but because they let you read the landscape clearly. A person who knows what a tracker pixel is can look at an email client differently. A person who understands what end-to-end encryption does — and does not — protect can make better choices about which messaging apps to use.

This is not a guide that promises to make you invisible online, because nothing will. It is a guide to the concepts, mechanisms, and tradeoffs that shape your privacy on a smartphone. Some of these concepts describe threats. Others describe protections. Several describe systems that are neutral in design but consequential in practice, depending on how companies choose to use them. All 20 are worth understanding if you carry a phone — which is to say, almost everyone.

The goal is not paranoia. It is clarity.

End-to-end encryption

Credit:  franckinjapan / Unsplash

End-to-end encryption is one of the most discussed and most misunderstood concepts in digital privacy. At its core, it describes a system in which a message is encrypted on the sender's device and can only be decrypted on the recipient's device. No one in between — not the app company, not the mobile network, not a government with a court order to the service provider — can read the content of the message in transit.

The "ends" in end-to-end encryption refer to the two endpoints of the communication: the sender and the recipient. In a non-end-to-end system, the service provider holds encryption keys and can decrypt messages as they pass through its servers. This is how most email works. The provider can read what you send, and so can anyone who gains access to the provider's infrastructure, whether through hacking, legal process, or insider access.

End-to-end encryption eliminates the service provider as a potential point of interception. WhatsApp, Signal, iMessage (between Apple $AAPL 0.70% devices), and several other messaging apps use end-to-end encryption for their messages and calls. This means that even if a government presents WhatsApp's parent company, Meta $META 0.27%, with a subpoena for the content of your messages, Meta cannot comply — it does not have the keys to decrypt them.

There are important limits to this protection. End-to-end encryption protects content in transit, but it does not protect metadata — the information about who sent a message, to whom, when, and how often. It also does not protect messages once they arrive on a device. If someone has physical access to your unlocked phone, the messages are readable regardless of how they were transmitted. Backups can also undermine the protection: if your messages are backed up unencrypted to a cloud service, that service can access them even if the messages were encrypted during transmission.

Some apps that claim to use end-to-end encryption have implemented it incompletely or incorrectly. The term is also sometimes applied loosely to describe encryption that is technically present but configured in a way that allows the provider to access content under certain conditions. Reading the actual technical documentation — or relying on audits conducted by independent security researchers — is more reliable than trusting marketing language.

Credit:  Brett Sayles / Pexels

Metadata is data about data. In the context of communications, it typically refers to the information that surrounds a message or call — who sent it, who received it, when it was sent, from what location, on what device, and for how long. It excludes the actual content of the communication.

This distinction is often used to minimize the significance of metadata collection. A common argument runs as follows: even if a government or company can see your metadata, they cannot see what you said, so your privacy is substantially intact. This argument is wrong. Metadata reveals patterns of behavior that are often more revealing than content.

Consider what can be inferred from call metadata alone. If you call a cancer treatment center, a bankruptcy attorney, and a marriage counselor within the same week, those three data points tell a story without any transcript. If your phone pings cell towers near a particular location every Thursday afternoon, a pattern emerges. If you call a specific person 40 times in a month and then never again, that says something too.

Security researchers and intelligence agencies have long understood that metadata analysis can be more powerful than content surveillance in some respects, precisely because it is amenable to automated processing at scale. Individual messages require human interpretation. Metadata can be processed algorithmically across millions of records simultaneously.

On smartphones, metadata is generated constantly and by many different systems. Your carrier records metadata about your calls and texts. Your apps record metadata about when you open them and what you do inside them. Your phone records metadata about the networks it connects to and the devices it encounters via Bluetooth. Location data is itself a form of metadata. All of this creates a detailed behavioral profile that can be assembled and analyzed by anyone who has access to the pieces — which includes data brokers, advertisers, app developers, and law enforcement.

The practical implication is that protecting content alone is insufficient. A message encrypted end-to-end still generates metadata. Choosing communication tools that minimize metadata collection — Signal, for example, is designed to retain as little metadata as possible — is a separate decision from choosing tools that encrypt message content.

Credit:  cottonbro studio  / Pexels

Your smartphone knows where you are, almost always. It determines this through a combination of GPS, Wi-Fi network triangulation, cell tower signals, and Bluetooth beacons. GPS provides the most accurate positioning but drains battery quickly. The other methods are less precise but use less power, so phones often combine them dynamically depending on circumstances.

Location data is among the most commercially valuable data that smartphones generate. It allows advertisers to serve ads tied to physical places — a coffee shop near where you're standing, a car dealership in the area you drove through last week. It allows retailers to analyze foot traffic patterns. It allows political campaigns to target people in specific precincts. It allows employers, insurance companies, and anyone else who can buy or access the data to draw inferences about behavior.

Location data is collected by apps in the foreground, while you're using them, and in the background, while you're not. The background collection is the part most users don't think about. An app that asked for location access a year ago — perhaps to show you nearby restaurants — may still be recording your location every few minutes, even when the app is closed.

On both Android and iOS, you can review which apps have location permissions and what level of access they have. iOS, in particular, offers granular control: you can give apps access to your location only while using the app, allow access at all times, or deny access entirely. You can also choose "precise" or "approximate" location, the latter of which reveals only a general area rather than a specific coordinate.

The broader problem is that location data, once collected, can be sold or shared in ways that are difficult to trace or control. Data brokers aggregate location data from many sources and sell it to third parties. Even if an individual app behaves responsibly, the ecosystem in which it participates may not. Several major telecommunications companies have been sanctioned by U.S. regulators for selling customer location data to third parties without adequate consent.

App permissions are the mechanism by which apps request access to the hardware and data on your device. When you install an app and open it for the first time, it may ask for access to your camera, microphone, contacts, photos, location, calendar, Bluetooth, or other resources. Granting permission gives the app the ability to access that resource until you revoke the permission.

The permissions system exists to create a layer of user control between apps and sensitive data. Before this system existed, apps on many platforms could access anything on a device without asking. The introduction of explicit permissions requests was a genuine improvement. But the system has significant weaknesses in practice.

One problem is that users tend to grant permissions without thinking carefully about them. The request appears at a moment when the user wants to accomplish something — open a camera, send a voice message — and clicking "allow" is the path of least resistance. The long-term implications of that permission, which may last for years, are not considered in the moment.

A second problem is that permissions requests are often poorly explained. An app might ask for access to your contacts "to help you find friends," without explaining that it will upload your entire contact list to its servers and retain it indefinitely. The explanation provided at the permission prompt is typically written by the app developer and is not independently verified.

A third problem is that permissions can be bundled or implied. Granting one permission may enable data collection that you didn't specifically authorize. An app with microphone access can theoretically detect ambient sounds, not just record what you intentionally say into it.

Reviewing your app permissions periodically — both iOS and Android provide settings menus that show all permissions, organized by app or by permission type — is one of the more practical steps a smartphone user can take. The presence of an app with camera or microphone access that you don't recognize or no longer use is worth investigating.

Credit:  George Diamanto /  Pexels

Data brokers are companies that collect personal information about individuals from many different sources,........

© Quartz