What Is a Keylogger? How It Works, Detection, Removal & Best Protection Tools

Every password you type, every private message you send and every card number you enter passes through your keyboard first, which is exactly why attackers, abusers and sometimes overreaching employers want to record it. A keylogger sits quietly between your fingers and the application you trust, and one captured login can unlock email, banking and company systems. The good news is that keylogging can usually be detected, contained and prevented once you understand how it works and what to do next.
TL;DR
A keylogger records keystrokes. It can be software, a small hardware device, or one feature inside a larger piece of malware. Authorized, disclosed monitoring exists, but secret use against another person is surveillance and may be a crime.
No single symptom proves a keylogger is present. Slowness, battery drain and pop-ups have many harmless causes, so use layered checks rather than one scan or one sign.
Detection and removal depend on your platform: Windows, macOS, Android, iPhone/iPad and Linux each have different built-in tools, and hardware keyloggers will not appear in a software scan.
If you suspect exposure, change passwords and sign out sessions from a different, trusted device. If the person watching you might be an abusive partner, put safety first and read the stalkerware guidance before changing anything.
Strong protection is layered: updates, reputable endpoint security, a password manager, passkeys or phishing-resistant multi-factor authentication, and careful software and device habits.
Businesses should treat stolen credentials as a likely follow-on to endpoint compromise, and test the web applications and APIs those credentials unlock. Application penetration testing does not remove a keylogger from a device.
Quick answer
A keylogger is software or hardware that records what you type, and sometimes screenshots or clipboard data, so someone else can read it. Authorized monitoring exists, but covert use steals passwords and messages. Protection combines updated operating systems, reputable endpoint security, password managers, passkeys or phishing-resistant multi-factor authentication, and careful software and device habits.

Protect the systems behind customer logins
Your customers trust a login page with their passwords. If your web app, SaaS platform or API has weak access controls, one stolen credential, whether it came from a keylogger or a phishing page, can reach far more data than it should.
Professionally scoped web application, API and SaaS application security assessments
Focus areas can include authentication, authorization, session management, API flaws and tenant isolation, depending on the agreed scope
Findings with remediation direction, plus retesting when separately agreed
Penetration testing examines your applications. It does not remove malware from a personal device. Scope, authorization and specialist availability are confirmed before any work begins.
Table of contents
What Is a Keylogger?
A keylogger, also called a keystroke logger, is a tool that records the keys pressed on a keyboard. Some versions also capture screenshots, clipboard contents or the name of the window you were typing in. The recorded data may stay on the device until someone collects it, or it may be sent to a remote server. The MITRE ATT&CK framework describes this technique, Input Capture: Keylogging (T1056.001), as recording keystrokes so credentials can be captured as they are typed.
Keylogging is a technique, not a single product. It can appear inside a trojan horse, a remote access tool, a spyware package or a wider malware family. That is why two statements are both true: not every keylogger is a virus, and not every virus logs keystrokes.
Legitimate monitoring versus abusive surveillance
Some keylogging is authorized. An employer may monitor company-owned devices under a written policy, a security team may use input logging during a controlled investigation, and a parent may want age-appropriate oversight. Parents who mainly want safer gaming can often start with built-in controls, which we cover in our guide to Roblox safety for kids. What separates these uses from abuse is authorization, disclosure and proportionality.
Secretly recording another adult's typing is a different matter. Laws on consent, wiretapping and unauthorized access differ by country and state, so this article cannot tell you what is lawful where you live. In many places, installing a keylogger on a device you do not own, or on a partner's phone without their knowledge, can lead to criminal charges. If you are unsure, ask a qualified lawyer before monitoring anyone.
Why a secure connection does not stop a keylogger
Websites use HTTPS to protect data while it travels between your device and a server. A keylogger works earlier. It captures the keys at your device, before the page encrypts anything. On an infected computer, a padlock icon in the browser does not make a typed password safe, which is why endpoint security and account protections both matter.
How Does a Keylogger Work?
At a high level, every keylogger performs four jobs. The details differ between tools, but the sequence is surprisingly consistent. This section stays defensive and explains only what you need to recognize and block the behavior.
Capture. The tool observes keyboard input. Software versions typically do this by hooking into the operating system's input handling, and MITRE lists approaches such as hooking the API callbacks that process keystroke data. Hardware versions sit in the signal path between the keyboard and the computer.
Store. Captured keys are written to a local file, a hidden buffer or the memory of a small hardware device. Many tools add timestamps and the name of the active program so the log is easier to read later.
Report. The log is collected. Software tools may send it by email, upload it to a server or push it to a cloud account. Hardware devices usually need someone to retrieve them or, in some products, connect over a short-range wireless link.
Persist. Malicious versions try to keep running after a restart. Persistence is one reason a reboot alone rarely fixes an infection, and why deeply hidden code such as a rootkit can make cleanup harder.
Because these steps happen quietly, you rarely see the capture itself. You see the consequences, such as an account login from a place you have never been. That is why account monitoring and device checks work best together.
Two limits are worth knowing. First, a keylogger sees what is typed or pasted on the infected device, not what you type on a different, clean one. Second, many modern credential thieves do not need keystrokes at all and instead pull saved passwords or session cookies from the browser. We compare these methods later in this guide.
Types of Keyloggers
Keyloggers are usually grouped by where they run and how they are installed. The table summarizes the main categories and what each means for detection.
Type | Where it runs | Needs physical access? | Typical detection route |
User-level software | Inside the logged-in user's session | No, often arrives through a download or email | Security software, process and startup review |
Kernel- or driver-level software | Deep in the operating system | Not usually | Offline or boot-time scans, expert analysis |
Browser-based | Malicious extension or injected page code | No | Extension review, browser reset, endpoint security |
Hardware inline device | Plugged between keyboard and computer | Yes | Physical inspection of cables and ports |
Modified or embedded hardware | Inside a keyboard or other peripheral | Yes | Hardware inspection, replacing the device |
Software keyloggers are the most common and the most varied. User-level tools are simpler and easier for security products to spot. Kernel-level tools are harder to find because they run with higher privileges, which is where specialist help or an offline scan can matter. Browser-based loggers read what you enter into web forms through a harmful extension or page script, and they never touch the rest of the system.
Hardware keyloggers are small devices or modified peripherals that record keystrokes without any software on the computer. Because they run outside the operating system, a software scan will not see them. Physical checks of cables, ports and unfamiliar adapters are the realistic defense, particularly on shared or public computers. If you want a refresher on the physical side of computing, our computer hardware guide explains the components involved.
Researchers have also studied side-channel approaches, such as inferring keystrokes from sound or motion. These are mostly relevant to targeted, specialist scenarios and are not the main risk for everyday users, so this guide focuses on the types above.
How Keyloggers Reach Devices
Most keyloggers arrive through ordinary routes. They rarely need advanced tricks, because people can be persuaded to install them or to open the door themselves.
Malicious downloads. Cracked software, fake installers, pirated tools and bundled "free" utilities are common carriers. Our guide to adware shows how unwanted bundles often arrive the same way.
Trojans and email attachments. A file that appears to be an invoice, a document or a game can install a logger in the background. Read more about trojan horses and how they disguise themselves.
Phishing and social engineering. A convincing message pushes you to open a link or run a file. Understanding social engineering is one of the most effective defenses, and phishing simulation software helps teams practice it safely.
Malicious browser extensions. Extensions can read what you type into pages, so install only what you need from reputable publishers and review permissions regularly.
Weak or exposed remote access. Reused passwords, open remote desktop services and unpatched systems give attackers a path in. A firewall and sensible network security reduce that exposure.
Physical access. Someone with a few minutes at an unlocked computer can plug in a hardware logger or install monitoring software. Shared offices, public terminals and abusive relationships are the usual settings.
Legitimate tools used the wrong way. Commercial monitoring software, sometimes marketed to parents or employers, can be installed secretly on another person's device. The US Federal Trade Commission treats this as stalkerware, software installed without the owner's knowledge.
Notice what is missing from this list: a requirement to be a high-value target. Automated campaigns send the same lure to millions of people, which is why basic hygiene beats hoping you are too small to matter.
What Information Can a Keylogger Capture?
A keylogger captures whatever is typed on the infected device while it is running. In practice that usually means the most valuable text a person enters in a day.
Credentials. Usernames and passwords for email, banking, social media, cloud services and work systems. MITRE notes that attackers often turn to keylogging when other credential-theft methods fail.
Private communication. Messages, emails, notes and anything else typed into chat apps and documents.
Financial details. Card numbers, bank logins and the answers to security questions typed into forms. Banks watch for unusual transactions, a theme we explore in our guide to AI fraud detection in banking.
Search and browsing intent. Search terms, addresses and form entries that reveal habits and plans.
Business information. Internal system logins, customer details, draft contracts and code comments.
What a keylogger may miss
Keyloggers are powerful, but they are not magic. A password manager that fills a login automatically means you never type that password, although some malware also reads the clipboard, the screen or the form data directly, so a manager is not a complete shield. Signing in with a passkey avoids typing a reusable secret altogether. A one-time code you type on an infected device can still be captured, but it expires quickly, which limits how long it is useful.
Just as important, a keylogger sees only the device it runs on. If you approve a login on a separate, clean phone, the logger on your laptop never sees that approval. That separation is the reason this guide keeps recommending a second trusted device for recovery.
Not all credential theft is keylogging
Many modern attacks skip keystrokes and take saved passwords, browser cookies and session tokens straight from the browser or the system. These are usually described as information stealers, and we compare them with keyloggers later. The practical lesson is that defending against typed-password theft is necessary but not sufficient, so protect stored credentials and sessions too.

When a stolen password meets weak access controls
A captured username and password is only the beginning. What an attacker can do next depends on how your application handles permissions, sessions and API access. Weak authorization can turn one ordinary account into a view of someone else's data.
A scoped review can cover authentication, authorization checks and session management in your web application
API access control and tenant isolation can be examined for SaaS platforms, depending on the agreed scope
Results are tied to the agreed scope, test window and rules of engagement, so no one can promise every weakness will be found
This service assesses your applications and APIs. It does not stop a keylogger on a laptop, so it works best alongside endpoint security and strong authentication.
Signs of a Possible Keylogger
Here is the honest starting point: most keylogger symptoms are not specific. Software loggers are designed to stay quiet, and the clues below also appear with ordinary slowdowns, aging hardware and unrelated software problems. Treat them as reasons to investigate, not as proof.
Device-level clues
Typing that lags, skips or repeats characters, especially if it started suddenly.
Programs, browser extensions or startup items you do not recognize.
Security software that was switched off, would not update, or keeps being blocked.
Unexpected network activity from a program you cannot identify.
Strange behavior only on one device, while your other devices are fine.
Account-level clues
Login alerts or sign-ins from places and devices you do not recognize.
Password-reset emails you did not request.
Contacts receiving messages you did not send.
New forwarding rules, recovery emails or phone numbers added to an account.
Hardware clues
On a desktop, check whether something is plugged between the keyboard and the computer, such as an unfamiliar adapter or small cylinder on a cable. A keyboard that looks different from the one you left, or one you did not buy, is worth a second look on a shared or public machine. A laptop's built-in keyboard is far harder to tamper with, so external devices deserve the closest inspection.
How to weigh the evidence
Account-level clues are often stronger than device-level clues, because they show that someone else used your credentials. A single slow startup tells you almost nothing. A security alert from a reputable product, an unknown startup entry and a suspicious login together tell you quite a lot. If you are seeing a mix and still unsure, move on to the platform checks below.
How to Detect a Keylogger on Windows, macOS, Android, iPhone/iPad and Linux
Detection means looking for evidence from several angles. No single scan, process name or "anti-keylogger checker" can prove a device is clean. Menu names also change between versions and manufacturers, so treat the paths below as a guide and check your own settings. The table gives a quick map, and the platform notes follow.
Platform | Where to look first | Built-in protection | Main limitation |
Windows | Windows Security scans, startup apps, installed apps, browser extensions | Microsoft Defender Antivirus, including an optional offline scan | Offline scan has prerequisites and is not available on every Windows type |
macOS | Privacy & Security permissions such as Input Monitoring | Apple's built-in protections and permission prompts | An app asking for access is not automatically malicious |
Android | Play Protect, accessibility and device-admin permissions | Google Play Protect | Menu names vary by manufacturer |
iPhone/iPad | Safety Check, configuration profiles, third-party keyboards | App review and sandboxing | No desktop-style antivirus scan exists |
Linux | Startup services, scheduled jobs, input-device access, package integrity | Distribution package signing and permissions | No single magic detection command |
Windows
Start with the basics, then escalate.
Make sure Windows and your security software are fully updated, then run a full scan from Windows Security under Virus & threat protection and Scan options.
Review Settings, then Apps, for programs you did not install, and open Task Manager's Startup tab for unfamiliar entries. Advanced users can inspect startup locations in more detail with Microsoft's Sysinternals Autoruns tool.
Remove browser extensions you do not recognize, then reset the browser if needed.
For stubborn or deeply hidden malware, consider a Microsoft Defender Offline scan. Microsoft's documentation says it runs outside the normal Windows environment and targets threats such as rootkits. It requires Microsoft Defender Antivirus to be the primary antivirus and Windows Recovery Environment to be enabled, it is not supported on ARM-based Windows or Windows Server, and it restarts the device, so save your work first. Suspend BitLocker before running it to avoid a recovery-key prompt.
Our overview of Windows explains the system features mentioned here, and our guide to endpoint protection explains how security software fits around them.
macOS
Apple lets you control which apps may watch your keyboard. In System Settings, open Privacy & Security and then Input Monitoring, as described in Apple's guidance on input monitoring, and review each app on the list. Do the same for Accessibility and Screen & System Audio Recording, because Apple's list of privacy settings says Input Monitoring allows apps to see what you type. Many legitimate tools, such as remote support software, window managers and some utilities, ask for these permissions, so a listed app is a prompt to check, not proof of malware. Remove access for anything you no longer use or cannot identify, and also review login items and any configuration profiles on a Mac you personally manage. Our guide to macOS covers the platform in more detail.
Android
Open the Google Play Store, tap your profile icon, choose Play Protect and then Settings, and confirm that scanning apps with Play Protect is on, as Google's Play Protect help page describes. Then look at which apps have powerful permissions. Accessibility services and device-administrator rights are commonly abused because they can read what appears on screen, so remove any you do not recognize. Menu names differ by manufacturer, and apps installed from outside Google Play deserve extra suspicion. Our Android OS guide explains the platform's permission model.
iPhone and iPad
Apple's closed design makes a traditional keylogger far less common, so the realistic problems are different: a shared Apple Account, a configuration profile you did not approve, a third-party keyboard, or someone with physical access to your device. Open Settings, then Privacy & Security and Safety Check, to review what you share and which devices are signed in, as Apple's Safety Check guide explains. It requires iOS 16 or later and an Apple Account with two-factor authentication. Also check Settings, General and VPN & Device Management for unfamiliar profiles, and review any third-party keyboards that have Full Access, since those can see what you type. There is no consumer "full antivirus scan" on iOS, so keep the system updated and be cautious about jailbreaking. Read more in our iOS overview.
Linux
Linux gives you more visibility but less hand-holding. Review enabled services and user-level startup entries, scheduled jobs, and autostart folders for anything unfamiliar. Check which accounts belong to the input group, since access to input devices is what a logger needs, and use standard tools to see which processes have input devices open. Verify installed packages with your distribution's integrity-check tools, and read system logs for unexpected changes. There is no single magic command, so if you cannot explain something you find, preserve evidence and escalate to a specialist. Our guides to Linux and the Linux kernel provide helpful background.
Hardware keyloggers on any platform
Disconnect the keyboard and inspect the cable path from keyboard to port, including any adapters. Replace a suspect keyboard with a known-good one from a sealed package, and use your own keyboard on shared computers when you can. On a workplace machine, report what you find to IT rather than removing it yourself, so the device can be examined.
When to stop and escalate
Bring in a professional if the device belongs to your employer, if money or sensitive data may already be lost, if scans keep finding something that returns, or if you suspect you were targeted. Consumer cleanup and expert incident investigation are different jobs, and deeper cases may need specialist threat detection and response capability.
Safety first if you may be monitored by someone close to you. The US Federal Trade Commission warns that removing stalkerware can alert the person who installed it, who may escalate abuse. It advises contacting a domestic violence advocate first, using a device that is not monitored, such as a friend's phone, to seek help, and documenting evidence before you reset anything. In the United States, the National Domestic Violence Hotline can be reached at 1-800-799-7233 or by texting START to 88788. If you are elsewhere, contact local domestic violence or victim-support services.
How to Remove a Keylogger Safely
Removal is safest when you separate three jobs: protecting your accounts, cleaning the device, and recovering afterward. Doing them in the wrong order can hand your new passwords straight to the logger.
Check for abuse first. If someone close to you may have installed monitoring software, do not remove it yet. Read the safety note in the detection section and speak with a domestic violence advocate before changing anything.
Use a different trusted device for account changes. Do not change important passwords on the suspect computer, because the logger can record the new ones. Use a clean phone or computer instead, as covered in the next section.
Contain the device. Disconnect it from the network if you can, and stop signing in to sensitive accounts on it.
Back up only what you need. Copy documents, photos and other personal files to external storage. Avoid backing up applications or system settings, because a backup can carry the problem with it. The FTC similarly advises against restoring apps from a backup after stalkerware. Our guide to backup as a service explains safer backup options for businesses.
Update and scan. Install operating system and security updates, then run a full scan with reputable security software and quarantine whatever it finds. On Windows, use an offline scan for stubborn cases.
Remove what you cannot trust. Uninstall unknown apps, delete unfamiliar browser extensions, and remove suspicious profiles or administrator permissions. For deeper infections, see our guides to removing a rootkit and removing a computer virus.
Rebuild when confidence is low. If a scan keeps finding the same threat, if security tools will not run, or if you simply cannot be sure, a clean reinstall or factory reset is the more reliable path. Reinstall apps from official sources and restore only the personal files you backed up.
Replace hardware loggers. Unplug and discard a suspect inline device, replace the keyboard, and report it to IT if the machine is shared or belongs to work.
Change credentials again. Once the device is clean, change important passwords again from that device, and keep using strong authentication.
If your security software will not install or is being blocked, try the vendor's offline or rescue tool, or ask a professional. Scanning is evidence, not proof. A clean result lowers risk but cannot guarantee the device is safe, especially against hardware devices or deeply hidden code.
What to Do After Suspected Exposure
Assume that anything typed on the suspect device may have been seen. The goal now is to make stolen data less useful. Work from a different, trusted device, ideally one you know is clean and updated.
Start with email. Your main email account can reset almost every other password, so secure it first with a new, unique password.
Sign out everywhere and revoke access. Use each service's security settings to end other sessions, remove unknown devices and revoke app passwords or connected apps. Sessions can keep working even after a password change on some services.
Review recovery methods and MFA. Remove unfamiliar recovery emails and phone numbers, check forwarding rules and mailbox filters, and re-enroll multi-factor authentication where needed.
Prioritize critical accounts. Banking, password manager, cloud storage, social media and work logins come next, with a different password for each.
Move to passkeys where offered. A passkey removes the reusable secret that a keylogger would otherwise steal.
Watch for fraud. Set up account alerts, review statements, and ask your bank about fraud alerts or other protections available in your country.
Tell the right people. If a work device or work account was involved, inform your IT or security team right away. Keep screenshots and alerts as evidence, and avoid wiping a work device before they advise you.
Assume reuse is a risk. If you reused a password anywhere, change it everywhere. Attackers try stolen credentials against many services.
Speed matters because credentials do not stay idle. Verizon's 2026 Data Breach Investigations Report, as summarized by Help Net Security, found that half of ransomware victims who had a prior credential leak were attacked within 95 days of the leak. For more on what follows a serious compromise, see our guides to ransomware and cyberattacks.
Best Keylogger Protection Tools Compared
No product on the market is a dedicated, independently benchmarked "keylogger detector," so this comparison uses a safer approach. It looks at endpoint security tools whose broad malware protection has been tested by independent labs, and it notes what each tool does not do. Broad malware scores are not keylogger-specific scores, and we have not tested any of these products ourselves.
For Windows, AV-TEST's home Windows results for July and August 2026 tested products on Windows 11 with default settings. Microsoft Defender Antivirus, Malwarebytes Premium, Bitdefender Total Security, ESET Security Ultimate and Norton 360 all received the Top Product award. They all scored 6 out of 6 for protection, and Defender scored 5.5 for performance. AV-Comparatives separately published its Real-World Protection Test covering February to May 2026, which evaluated 20 consumer products, including these vendors, against web-based threats.
Option | Category and platforms | Independent evidence | Main limitation | Best fit |
Built-in endpoint protection, Windows | AV-TEST, Aug 2026: protection 6, performance 5.5, usability 6 | Windows only; offline scan has prerequisites | Windows users who want a solid built-in baseline | |
Anti-malware for PCs, Macs, Android and iPhone; a free option exists | AV-TEST, Aug 2026: 6, 6, 6 (Premium) | Broad malware focus, not a keylogger-specific guarantee | People who want a second opinion scanner or a simple suite | |
Consumer suite, Windows, macOS, Android, iOS | AV-TEST, Aug 2026: 6, 6, 6 | Paid suite; extra features may not suit everyone | Households with mixed devices | |
Consumer antivirus, Windows, macOS, Android | AV-TEST, Aug 2026: 6, 6, 6 (Security Ultimate) | Home iOS support is not listed on the home page | Users who prefer a light, configurable tool | |
Consumer suite, Windows, macOS, Android, iOS | AV-TEST, Aug 2026: 6, 6, 6 | Paid suite with bundled extras | Families wanting one multi-device plan | |
Built-in Android app scanning | No comparable lab figure used here | Not a full security suite; limits vary by device | Every Android user, as a baseline | |
Apple built-in protections | App review, sandboxing and permission prompts on iPhone, iPad and Mac | Not covered by the lab results above | Cannot catch every social-engineering or permission abuse | Apple users who keep systems updated |
How to choose
Start with what you already have. On Windows, Defender is a credible baseline, and the lab results show that paid suites are not automatically better at protection. Add a paid product when you need coverage across several devices, a family plan or extra features such as a VPN, which are not the same as keylogger protection. A reputable second-opinion scanner is a reasonable add-on for on-demand checks, although running multiple real-time antivirus programs together can cause conflicts.
Whatever you pick, remember three limits. Software cannot see hardware keyloggers. A scan cannot prove a device is clean. And endpoint tools protect the device, not the logins your business exposes to the internet. Enterprise buyers should look at managed detection and response, and our guides to AI endpoint protection and AI threat detection cover that category. We do not use affiliate links or discount codes, and prices change often, so check each vendor's current terms.

Endpoint security and application testing solve different problems
Antivirus and endpoint tools protect the devices your team types on. Application and API penetration testing examines the systems those devices sign in to. A company that covers only one side leaves the other open, which is why they work together rather than replace each other.
Endpoint security aims to block or detect malware on laptops, phones and servers
A scoped application assessment looks for weaknesses in your web app, SaaS platform or API that a stolen login could exploit
Retesting after fixes can be agreed separately to confirm that issues were addressed
An assessment is scoped to what you agree and authorize. It does not guarantee that every issue will be found, and it is not a substitute for endpoint protection.
How to Prevent Keyloggers and Credential Theft
Prevention is about layers. Each layer is imperfect, but together they make stolen keystrokes far less useful.
Keep everything updated. Operating systems, browsers and apps close the holes that malware uses. Our software maintenance cost guide explains why patching is worth budgeting for.
Use standard accounts day to day. Reserve administrator rights for installs, since malware with full rights can do more.
Install from trusted sources. Avoid cracked software, unknown installers and unneeded browser extensions.
Slow down on messages. Treat unexpected links and attachments with suspicion, and verify requests through another channel.
Run reputable endpoint security and keep real-time protection on.
Use a password manager. Unique passwords limit damage, and a good manager only autofills on the correct site, which can reveal fake login pages.
Adopt passkeys or phishing-resistant MFA. The US Cybersecurity and Infrastructure Security Agency, in its Implementing Phishing-Resistant MFA fact sheet from October 2022, urges organizations to adopt phishing-resistant MFA and names FIDO/WebAuthn as the most widely available option. These methods reduce what a stolen password can do, but they cannot make an infected device trustworthy.
Secure physical access. Lock screens, restrict who can plug devices into workstations, and avoid typing sensitive logins on public computers.
Keep backups. Recent, clean backups make a rebuild painless.
What about on-screen keyboards and virtual typing tools?
On-screen keyboards can avoid some basic hardware and keystroke-based loggers, but they do not stop malware that records the screen or reads other input. Treat them as a minor aid, not a substitute for the layers above.
For organizations
Teams should add written monitoring policies, device management, USB and application controls, staff training and central logging. A SIEM helps correlate alerts, a firewall and network security controls limit lateral movement, and cloud security practices protect the accounts keyloggers aim to unlock. EU organizations should also review NIS2 obligations, and our overview of cybersecurity services shows how to evaluate outside help.
Real-World Keylogger Cases and Myths vs. Facts
Documented cases show how keyloggers are really used, and how ordinary the weaknesses behind them can be. The three below come from public law-enforcement and news records. Charges described as allegations remain allegations unless a court has decided otherwise.
Case 1: A public-computer password harvest at Kinko's (2001 to 2002)
According to the US Department of Justice's sentencing announcement, Juju Jiang admitted that between February 2001 and December 2002 he installed keylogging software on computers at Kinko's stores in Manhattan to collect customers' usernames and passwords. He was sentenced to 27 months in prison. The Associated Press reported that he captured more than 450 logins. Lesson: avoid signing in to sensitive accounts on shared or public computers.
Case 2: The attempted £220 million Sumitomo Mitsui bank theft (2004 to 2005)
UK police announced in March 2005 that they had foiled an attempt to steal about £220 million from the London offices of Sumitomo Mitsui, as InformationWeek reported. Press reports said the plot involved keylogging devices, and The Register reported that investigators were examining claims that cleaning staff had attached hardware devices to keyboard sockets. The bank said it suffered no losses. Lesson: physical access and hardware loggers are real risks that software scans cannot see.
Case 3: Warzone RAT sold as a service (2024)
The US Department of Justice announced in February 2024 that it had disrupted the Warzone RAT malware service, which it said could record keystrokes, steal usernames and passwords, take screenshots and activate webcams. Two people, in Malta and Nigeria, were charged in connection with selling it and supporting buyers. Lesson: keylogging is often sold as one feature in affordable, supported malware, so the barrier to entry is low.
Myths versus facts
Myth | Fact |
A keylogger is always a virus. | Keylogging is a technique. It can be authorized monitoring, a stand-alone tool or one feature of a larger malware family. |
A clean scan proves my device is safe. | A scan lowers risk but cannot prove safety, and software cannot see hardware loggers. |
The HTTPS padlock stops keyloggers. | HTTPS protects data in transit. A keylogger captures keys at the device before encryption. |
Multi-factor authentication makes keyloggers harmless. | It helps, but typed one-time codes can still be captured, and an infected device cannot be trusted. Phishing-resistant methods are stronger. |
Only companies and celebrities are targeted. | Automated campaigns send lures to huge numbers of people, so ordinary users are exposed. |
Keylogger vs. Spyware, Infostealers, Rootkits and Related Threats
These terms overlap, which causes confusion. A good rule is that a keylogger describes what is collected, while the other terms describe how the software behaves or hides.
Term | What it means | Relationship to keylogging |
Keylogger | Records keystrokes | The core technique |
Software that secretly gathers information about activity | May include keylogging among other monitoring | |
Stalkerware | Spyware installed on someone's device without their knowledge to watch them | Often logs typing along with location, messages and more |
Infostealer | Malware that collects saved credentials, cookies and files | May or may not log keys, and often grabs stored data instead |
Remote access trojan | Malware giving an attacker remote control | Frequently includes keystroke recording, as in Warzone |
Code that hides the presence of other software | Can conceal a keylogger from ordinary scans | |
Malware disguised as something harmless | A common delivery vehicle for loggers |
Infostealers deserve special mention because they bypass typing altogether. In October 2024, an international operation led by Dutch police seized infrastructure tied to the RedLine and META infostealers, and the US Department of Justice announced its participation. For broader context, see our guides to malware, computer worms and AI-powered malware.
Keylogger Risks for Businesses and When Application Penetration Testing Helps
For a business, a keylogger is rarely the end of the story. It is often the first step toward using valid credentials against the systems customers and staff rely on. Verizon's 2026 Data Breach Investigations Report, as reported by Descope, found vulnerability exploitation at 31 percent of initial access, ahead of credential abuse at 13 percent. But GitGuardian's analysis of the same report notes that credential abuse appears in 39 percent of breaches when you follow the full attack. In other words, weak applications and stolen logins reinforce each other.
That is where application penetration testing can help. Once an attacker holds a valid login, the question becomes what that login can reach. A scoped assessment of a web application, SaaS platform or API can examine authentication, authorization, session handling and tenant isolation, which determine whether one compromised account stays contained. Our guide to penetration testing services explains the types and process, and our overviews of custom web application development, client portals and APIs show where these controls matter. Unpatched servers add another path, as the PHP2Shell vulnerability shows, and broader strategy is covered in our guide to cybersecurity infrastructure and the cyber pandemic.
What penetration testing does not do
Be clear about the limits. Articsledge's penetration testing is focused on web application, API and SaaS application security, and it is not antivirus, keylogger removal, device forensics or incident response. It does not neutralize a keylogger on a laptop or undo stolen credentials. Businesses still need endpoint security, strong authentication, monitoring and a response plan. Testing complements those controls by checking the applications behind them, and it works only within an agreed scope and authorization.

Turn security uncertainty into a prioritized plan
If you run a customer-facing application and are unsure how far a stolen login could reach, a scoped assessment gives you something concrete to act on rather than a vague worry.
Authorized scoping so the work targets the systems you actually want assessed
Practical findings with prioritized remediation direction
Retesting after fixes, when separately agreed
Please do not send passwords, API keys or vulnerability details in your first message. Share the scope and goals, and the details can be handled securely after authorization is agreed.
Frequently Asked Questions
Is it legal to use a keylogger?
It depends on who owns the device, who consented and where you live. Authorized, disclosed monitoring of your own or company-owned devices can be lawful. Secretly recording another adult's typing is illegal in many places. This is general information, not legal advice, so ask a qualified lawyer before monitoring anyone.
Is a keylogger a virus?
Not necessarily. A keylogger is a function that records typing. It can be a stand-alone tool, a feature of spyware or a remote access trojan, or an authorized monitoring product. A virus is a self-replicating program, and most keyloggers do not replicate.
Can antivirus detect every keylogger?
No. Security software can catch many known and suspicious loggers, but new, custom or deeply hidden ones may slip through, and hardware loggers are invisible to it. A clean scan lowers risk without proving safety, so combine scans with account monitoring and physical checks.
Can a keylogger steal my passwords?
Yes, if you type them on the infected device. A logger records the keys before a website encrypts anything. Password managers, passkeys and separate trusted devices reduce the exposure, but other malware may still read saved data or sessions.
Can software detect a hardware keylogger?
Generally not, because the device sits outside the operating system. Inspect cables, ports and adapters between the keyboard and computer, replace suspect keyboards, and avoid typing sensitive logins on shared computers. On work machines, report anything unusual to IT.
Can a keylogger work offline?
Yes. Many store keystrokes locally and send them later, and hardware devices store them on the device itself. An internet connection is only needed for remote collection, which is why unplugging a network cable does not make a logged keystroke disappear.
Can an iPhone or iPad have a keylogger?
It is uncommon, thanks to Apple's app review and sandboxing. More realistic risks are shared accounts, unfamiliar configuration profiles, third-party keyboards with Full Access, jailbreaking and physical access. Use Safety Check, review profiles and keyboards, and keep iOS updated.
How do I check for a keylogger on a Mac?
Open System Settings, then Privacy & Security, and review Input Monitoring, Accessibility and Screen & System Audio Recording for apps you do not recognize. Many legitimate tools request these permissions, so research an app before removing it, then run reputable security software.
Will a factory reset remove a keylogger?
Usually, for software loggers, because it erases the system and apps. Restore only personal files, not apps or system backups, and reinstall software from official sources. A reset will not remove a hardware keylogger, and some firmware-level threats need expert help.
Do MFA and passkeys stop keyloggers?
They reduce the damage but do not cure infection. A passkey avoids typing a reusable secret, and phishing-resistant MFA is harder to replay. Codes typed on an infected device can still be captured, and the device itself remains untrusted until cleaned.
Are free tools enough?
For many people, built-in protection such as Microsoft Defender or Google Play Protect, plus updates and good habits, is a sensible baseline. Paid suites add multi-device coverage and extras. Independent lab results show strong protection from both free and paid tools.
What should I do if I think my passwords were stolen?
Use a different trusted device. Secure your email first, change passwords, sign out other sessions, check recovery options and MFA, and watch your accounts. Then clean the original device. If an abusive person may be monitoring you, seek safety advice before changing anything.
Is consumer keylogger cleanup the same as business penetration testing?
No. Cleanup removes malware from a device. Penetration testing assesses authorized web applications, APIs and SaaS systems for weaknesses. Articsledge's service covers the second job and does not remove keyloggers, recover stolen credentials or provide incident response.
Key Takeaways
A keylogger is a technique for recording keystrokes. It can be authorized monitoring, a stand-alone tool or one feature inside other malware.
Symptoms are weak evidence, so combine device checks, account activity and reputable scans before drawing conclusions.
Hardware keyloggers are invisible to software, so inspect cables, ports and shared computers physically.
Recover from a different, trusted device, starting with email, and sign out other sessions after changing passwords.
If someone close to you may be monitoring you, put safety first, because removing stalkerware can alert them.
Layered protection works best: updates, endpoint security, a password manager, passkeys or phishing-resistant MFA, and careful habits.
Businesses should pair endpoint protection with application and API testing, remembering that testing does not remove a keylogger from a device.
Actionable Next Steps
If you may be monitored by someone close to you: stop and contact a domestic violence advocate or local support service from a device that is not monitored before you change or remove anything.
Contain the device: limit sensitive use, disconnect it from the network where possible, and avoid typing new passwords on it.
Recover accounts from a trusted device: secure email first, then banking, your password manager, cloud and work accounts, signing out other sessions and checking recovery options.
Check and clean the device: update the system, run a full scan, review startup items, extensions and permissions, and consider an offline scan on Windows.
Rebuild when unsure: back up personal files only, reset or reinstall, and reinstall apps from official sources.
Inspect hardware: check keyboard cables, ports and adapters, and replace anything suspicious.
Strengthen defenses: turn on automatic updates, use a password manager, and move to passkeys or phishing-resistant MFA wherever you can.
Escalate in business settings: tell IT or security, preserve evidence, and consider a scoped application and API assessment to check what stolen credentials could reach.
Glossary
Keylogger: software or hardware that records keystrokes.
Hardware keylogger: a device or modified peripheral that records typing without software on the computer.
Spyware: software that secretly gathers information about a user's activity.
Stalkerware: spyware installed on someone's device without their knowledge to monitor them.
Infostealer: malware that collects saved credentials, cookies and files.
Remote access trojan (RAT): malware that lets an attacker control a device remotely.
Rootkit: code designed to hide the presence of other software.
Endpoint security: protection for devices such as laptops, phones and servers.
Multi-factor authentication (MFA): requiring more than one proof of identity to sign in.
Passkey: a phishing-resistant sign-in method based on cryptographic keys instead of a typed password.
Session token: a value that keeps you signed in, which attackers may try to steal or reuse.
Penetration testing: an authorized, scoped test that looks for exploitable weaknesses in a system or application.
Sources & References
Input Capture: Keylogging (T1056.001), MITRE ATT&CK, last modified 12 May 2026.
Microsoft Defender Offline scan, Microsoft Learn, last updated 3 July 2026.
Stalkerware: What To Know, US Federal Trade Commission, last modified 10 August 2026.
Google Play Protect help, Google Play Help, n.d.
Control access to input monitoring on Mac, Apple Support, n.d.
App privacy access settings, Apple Platform Support, n.d.
Safety Check on iPhone, Apple Support, September 2026.
Home Windows antivirus results, AV-TEST, results listed August 2026.
Real-World Protection Test, February to May 2026, AV-Comparatives, June 2026.
Implementing Phishing-Resistant MFA, CISA, October 2022.
Verizon DBIR: Vulnerability exploitation is the dominant initial access vector, Help Net Security, 20 May 2026.
Verizon DBIR 2026 analysis, Descope, n.d.
Initial access changed, the attack path did not, GitGuardian, n.d.
Queens Man Sentenced to 27 Months' Imprisonment, US Department of Justice, 28 February 2005.
Keyloggers Foiled In Attempted $423 Million Bank Heist, InformationWeek, n.d.
Mission Impossible at the Sumitomo Bank, The Register, 13 April 2005.
International Cybercrime Malware Service Dismantled, US Department of Justice, 9 February 2024.
US joins international action against RedLine and META infostealers, US Department of Justice, October 2024.
Vendor pages used for platform support, all n.d.: Windows Security, Malwarebytes, Bitdefender Total Security, ESET Home and Norton 360.


