Cybernion
← Insights

Security

What Is a Trojan Horse in Cyber Security?

4 Nov 2026 · Updated 4 Nov 2026

A trojan is malware disguised as legitimate software. It does not self replicate like a virus or worm, it relies on tricking a person into installing it, then delivers its real payload once running, commonly a backdoor, credential theft, or a downloader that fetches further malware.

The name comes from the wooden horse used to smuggle Greek soldiers into Troy, and the mechanism is the same idea in software form: something that looks harmless or useful on the outside conceals something harmful once let in.

What is a trojan horse

A trojan is a category of malware defined by how it gets onto a system, not by what it does once there. It presents itself as a legitimate file, application, or update. The victim chooses to run it, believing it is something else, or something safe. Once executed, the trojan carries out its actual function, which can range from opening a backdoor for remote access, to stealing credentials, to quietly downloading further malicious tools.

This distinguishes a trojan from other malware categories. A virus needs to attach to and infect other files to spread. A worm spreads on its own across a network by exploiting vulnerabilities, with no user interaction needed. A trojan does neither. It has no built in mechanism to spread itself, it depends entirely on deception and, usually, one deliberate action by the victim.

How do trojans work

The lifecycle of a trojan typically has three stages. First, delivery, the trojan is packaged to look like something the target wants or expects, an invoice, a cracked version of paid software, a game, a document, or a fake system update. Second, execution, the victim runs the file, and the trojan installs itself, frequently establishing persistence so it survives a reboot and evading whatever security software is present. Third, the payload activates, which is where trojans differ most from each other.

Some trojans are built to be quiet and long lived, maintaining access for as long as possible without drawing attention. Others act quickly, harvesting credentials or financial details in the first session and exfiltrating them immediately. A significant share of modern trojans function primarily as a delivery mechanism, opening the door for a second stage payload such as ransomware, a rootkit, or additional spyware, rather than doing the damage themselves.

What are the common types of trojans

Remote access trojans (RATs). Give the attacker ongoing, often full, control over the infected device, including file access, webcam and microphone access on some variants, and the ability to run further commands.

Banking trojans. Target online banking sessions specifically, often by injecting fake fields into legitimate banking websites in the victim’s browser, or by intercepting one time codes, to steal credentials and authorise fraudulent transactions.

Downloader trojans. Have a narrow job, fetch and install additional malware after the initial infection. They are often small and designed to avoid detection long enough to deliver a more capable payload.

Backdoor trojans. Create a persistent, hidden access point into the system that the attacker can use later, independent of the original infection vector.

Fake antivirus or scareware trojans. Present a fake warning that the system is infected and prompt the victim to pay for or install a fraudulent security tool, which is itself the trojan’s actual payload.

How do trojans reach victims

Email remains the most common delivery channel. A message disguised as an invoice, delivery notification, or official communication carries an attachment that is, or contains, a trojan. Fake software download sites and cracked or pirated software are a persistent source, users seeking a free version of paid software are a reliable target because they are already primed to bypass normal verification steps. Malicious browser extensions that request excessive permissions can act as trojans once installed. On mobile devices, apps distributed outside official app stores, or apps that request permissions unrelated to their stated purpose, are a common trojan vector.

Phishing and social engineering underpin almost every trojan delivery method, because the core requirement is convincing a person to trust and run something they should not.

How do we defend against trojans

Application control is the most direct technical defence. If only pre-approved software is permitted to execute, a trojan that a user is tricked into downloading generally cannot run, regardless of how convincing the disguise is. This is one of the ASD’s Essential Eight strategies and is specifically effective against this threat category.

User awareness matters because trojans depend on deception rather than a technical vulnerability. Training staff to verify unexpected attachments, avoid unofficial software sources, and check app permissions before installing reduces the number of opportunities a trojan has to succeed. Keeping software and operating systems patched limits what a trojan can do even if it does execute, since many trojans rely on further exploitation to gain elevated privileges. Endpoint detection and response (EDR) tooling helps identify trojan behaviour after installation, such as unexpected outbound connections to a command and control server, which is often the first visible sign once a trojan is active.

Frequently asked questions

What is a trojan horse in cyber security?

A trojan, or trojan horse, is malware disguised as or hidden inside legitimate looking software. Unlike a virus or a worm, a trojan does not self replicate. It relies on the victim voluntarily downloading, installing or running it, believing it to be a genuine application, document, or update, after which it delivers its actual malicious payload.

How is a trojan different from a virus or a worm?

A virus attaches to a host file and needs that file to run to spread. A worm self replicates and spreads across networks with no user action required. A trojan does neither, it does not attach to other files and does not self replicate. It relies entirely on social engineering to get a person to install it, then acts once it is running.

What are common types of trojans?

Common variants include remote access trojans that give an attacker ongoing control of the infected device, banking trojans that intercept online banking credentials and sessions, downloader trojans that fetch further malware after installation, and backdoor trojans that create a persistent unauthorised access point. Many modern trojans combine several of these functions in one payload.

How do trojans usually reach a victim?

Trojans typically arrive as email attachments disguised as invoices or documents, links to fake software download pages, cracked or pirated software from unofficial sources, malicious browser extensions, or fake mobile apps distributed outside official app stores. The common factor is that the victim is convinced to actively install something that looks legitimate.

How can organisations defend against trojans?

Application control that only allows approved software to run is one of the most effective defences, since a trojan generally cannot deliver its payload if it cannot execute. User education on phishing and unofficial software sources reduces the chance of installation in the first place, and endpoint detection and response tooling can catch trojan behaviour, such as unusual outbound connections, after installation.

Sources:

  1. Malware, Australian Signals Directorate, 2026
  2. Trojan horse, National Institute of Standards and Technology, 2026
  3. Essential Eight, Australian Signals Directorate, 2026

Last updated: 30 August, 2026

Talk to us about your engagement

A scoped proposal within one business day.