Security Controls, the CIA Triad & Zero Trust

Security+ SY0-701 · General Security Concepts

By Abu Anns · Last updated 20 November 2026Exam version: SY0-701

What a security control is

A security control is any safeguard put in place to protect an organization's assets — its data, systems, and people — from threats. Controls are the building blocks of every security program, and the SY0-701 exam expects you to classify any given control along two independent axes: its category and its type. These are separate questions about the same control, and mixing them up is one of the most common mistakes on the exam. A firewall, for example, is a technical control by category and a preventive control by type. A security guard is a physical control by category and a deterrent control by type. Always ask both questions: how is it implemented (category), and what is it meant to do (type)?

Control categories: how the control is implemented

There are four categories. Technical (sometimes called logical) controls are implemented in hardware or software: firewalls, encryption, access control lists, antivirus software, and intrusion detection systems. Managerial controls are the policies, procedures, and governance decisions that direct the security program: the acceptable use policy, risk assessments, security awareness training, and the decision to require background checks. Operational controls are the day-to-day human processes that keep security working: the person who reviews logs, the team that runs incident response, the guard who checks badges at the door. Physical controls are the tangible barriers and devices that protect facilities and hardware: locks, fences, bollards, security cameras, and badge readers.

Note the overlap: a security guard is a physical control, but the guard's patrol routine is an operational control, and the policy that says guards must patrol is a managerial control. The exam usually makes the intended category clear from the wording, but when a scenario is ambiguous, look for the strongest signal: if it is a device or a piece of software, it is technical; if it is a written rule, it is managerial; if it is a person doing a task, it is operational; if it is a barrier or a lock, it is physical.

Control types: what the control is meant to do

There are six types. Preventive controls stop an incident before it happens: a firewall blocking traffic, a locked door, encryption. Deterrent controls discourage an attack by making it seem less attractive: warning signs, visible cameras, a banner that says activity is monitored. Detective controls identify an incident while it is happening or after it has happened: intrusion detection systems, log monitoring, audits, motion sensors. Corrective controls fix the damage after an incident: restoring from backup, applying a patch, terminating a compromised session. Compensating controls are alternatives put in place when the ideal control cannot be used: if a legacy system cannot be patched, you might isolate it on its own network segment instead. Directive controls tell people what to do: policies, procedures, and standards that direct behavior.

The classic exam trap is confusing deterrent with preventive. A warning sign does not physically stop anyone — it only discourages, so it is a deterrent control. A fence physically stops entry, so it is preventive. Another trap is detective versus corrective: an alarm that sounds when a door opens is detective; the guard who responds and locks the door again is corrective. When a question asks for the type, focus on the moment the control acts: before the incident (preventive or deterrent), during or just after (detective), or in the recovery phase (corrective).

The CIA triad and the AAA framework

The CIA triad is the foundation of security thinking: confidentiality (only authorized people can see the data), integrity (the data has not been altered or tampered with), and availability (authorized users can reach the data and systems when they need them). Every security decision is a balance between these three. Encrypting a database protects confidentiality; hashing a file lets you verify its integrity; redundant servers and backups protect availability. A denial-of-service attack targets availability; data theft targets confidentiality; unauthorized modification targets integrity. Non-repudiation is often listed alongside the triad: it means a person cannot later deny having performed an action, and it is achieved with digital signatures and audit logs.

The AAA framework describes how access works: authentication (proving who you are, with a password, a token, or biometrics), authorization (deciding what you are allowed to do, based on permissions and policies), and accounting (recording what you did, through logs and audit trails). Authentication answers "who are you?", authorization answers "what can you do?", and accounting answers "what did you do?". The exam tests these distinctions constantly: a login prompt is authentication, a permission error is authorization, and a log entry is accounting.

Zero trust

Zero trust is a security model built on one principle: never trust, always verify. In a traditional network, anything inside the perimeter was trusted; zero trust assumes the network is already compromised and verifies every request, from every user and device, every time. The model has a control plane, which makes access decisions, and a data plane, which carries the actual traffic. Inside the control plane sit three components the exam names: the policy engine (which decides whether access is granted, using identity, device health, and context), the policy administrator (which carries out the decision by creating or removing access), and the policy enforcement point (which sits in the path of the traffic and allows or blocks it). Zero trust also relies on microsegmentation, continuous monitoring, and the assumption that no user or device is trusted by default, even if it is on the corporate network.

Deception technology and change management

Deception technology turns the attacker's own curiosity against them. A honeypot is a decoy system that looks valuable but exists only to attract and observe attackers; a honeynet is a whole network of honeypots; a honeyfile is a tempting fake document, and a honeytoken is a fake credential or piece of data (such as a bogus API key) that triggers an alert if anyone uses it. Because legitimate users have no reason to touch these decoys, any interaction is a strong signal of malicious activity.

Change management is the process that keeps security from being broken by well-meaning updates. A proper change process includes an approval step, a named owner, stakeholder review, an impact analysis, test results, a backout plan (how to undo the change if it fails), a scheduled maintenance window, and documentation with version control. The exam also tests the technical implications of change: allow and deny lists that must be updated, restricted activities during the change, downtime, system restarts, and dependencies on legacy applications. The classic scenario is a change that went wrong because it skipped testing or had no backout plan — and the exam's correct answer is almost always "follow the change management process" rather than "make the change faster."

Advertisement

Exam traps to watch for

Practice questions

Test yourself on this guide. Answers are checked instantly, with a full explanation for each question.

Q1.A company installs a firewall that blocks all inbound traffic except traffic on ports 443 and 25. Which control category and control type does the firewall represent?

Report an error

Q2.A data center posts large signs at every entrance warning that the facility is under 24-hour video surveillance and that intruders will be prosecuted. What type of control are the signs?

Report an error

Q3.A hospital runs an old medical imaging system that the vendor no longer supports, so it cannot be patched. The security team moves the system onto an isolated network segment with no internet access and restricts which workstations can reach it. Which type of control is the network isolation?

Report an error

Q4.An attacker floods a company's web server with traffic until legitimate customers cannot load the website. Which element of the CIA triad has been attacked?

Report an error

Q5.An employee successfully logs in to the company portal with a username and password, but when she tries to open the payroll folder she receives an 'access denied' message. Which part of the AAA framework is responsible for the denial?

Report an error

Q6.A company adopts a zero trust model. A laptop that is already connected to the corporate office network requests access to a finance application. Under zero trust, how is this request handled?

Report an error

Q7.A security team places a document named 'Q4_Bonus_Salaries.xlsx' on a shared drive. The file contains fake data, and opening it triggers a silent alert to the security team. What is this an example of?

Report an error

Q8.A network engineer plans to upgrade the firmware on all core switches during a weekend maintenance window. Before the change is approved, the change advisory board asks for a documented plan describing how to return the switches to their previous firmware if the upgrade fails. What is this plan called?

Report an error