A threat is something that could cause harm. A vulnerability is a weakness that lets it happen. Risk is what results when a real threat meets an exploitable vulnerability and causes an impact worth measuring. Confuse the three and the wrong thing gets patched, or a number nobody can act on gets reported.
Most security incident reports use these three words loosely, and most vendor pitches lean on the confusion to make every finding sound urgent. They are not interchangeable. Getting them right changes what gets fixed first, what is accepted, and what is reported to a board.
What is a threat in cyber security?
A threat is a potential source of harm, not the harm itself. It can be a person (a ransomware group, an insider), a piece of code (a malware family), or an event (a natural disaster, a power outage). A threat exists whether or not an organisation is vulnerable to it. Ransomware groups target Australian mid sized businesses regardless of whether any particular business has a patch backlog. That ongoing intent and capability is the threat.
Threats are usually described by capability and intent: what can this actor do, and do they want to do it to a given organisation. The Australian Signals Directorate’s Annual Cyber Threat Report and sector specific threat intelligence describe the current threat environment at a national level, but a threat only becomes relevant to one organisation when it lines up against something that organisation is actually exposed to.
What is a vulnerability in cyber security?
A vulnerability is a weakness in a system, process or person that a threat can exploit. It might be a missing patch, a misconfigured cloud storage bucket, a default password, a gap in a change control process, or a staff member who has not had phishing awareness training. Vulnerabilities are usually discoverable and fixable on their own terms, independent of whether any specific threat is currently targeting them.
The Common Vulnerabilities and Exposures (CVE) system and its severity scoring, the Common Vulnerability Scoring System (CVSS), catalogue technical vulnerabilities and rank how severe they are if exploited. See what CVSS scores mean for how that scoring works. A vulnerability scan or penetration test is how most organisations find these weaknesses before an attacker does.
What is risk, and how is it calculated?
Risk is the combination of threat, vulnerability and impact. The common formulation is risk equals likelihood multiplied by consequence, where likelihood already accounts for whether a real threat exists and whether the vulnerability is exploitable, and consequence accounts for what happens if it is. A critical vulnerability on an internet facing system that processes customer data is high risk. The same vulnerability on an isolated test system with no data is low risk, even though the vulnerability itself is identical.
This is why a raw CVSS score is not a risk score. CVSS measures the severity of a flaw in isolation. Risk asks whether a threat can reach that flaw, and what happens to the organisation if it does. Two organisations running the same vulnerable software can have very different risk ratings for it.
How do threat, vulnerability and risk work together?
They form a chain, and breaking any link removes the risk. A threat actor needs a vulnerability to exploit before they can cause an impact. Close the vulnerability and the threat has nothing to work with, at least through that path. Reduce exposure to the threat, through network segmentation or by taking a system offline, and the vulnerability becomes unreachable even if it stays unpatched. Reduce the impact, through backups or multi factor authentication on the next system in the chain, and the same combination of threat and vulnerability does less damage.
Most control frameworks work on more than one link at once. Essential Eight patching controls close vulnerabilities. Network segmentation and email filtering reduce exposure to the threat. Backups and incident response reduce impact once something does get through.
| Term | What it is | Example | Who controls it |
|---|---|---|---|
| Threat | A potential source of harm | A ransomware group, a malicious insider, a storm | Largely outside the organisation’s control |
| Vulnerability | A weakness a threat can exploit | An unpatched server, a weak password, an untrained user | Directly within the organisation’s control |
| Risk | Threat times vulnerability times impact | A ransomware group encrypting an unpatched, internet facing finance system | Managed through treatment, not eliminated |
Why do security teams confuse these terms, and why does it matter?
Because vendors and headlines use “risk” to mean “vulnerability found” and “threat” to mean “risk exists,” and the shortcuts save words. The cost shows up in prioritisation. A vulnerability scan that reports two hundred findings as “risks” gives a board nothing to act on, because it has not weighed likelihood or impact for any of them. A proper risk assessment might reduce that list to a dozen items that actually matter, ranked in an order that reflects what could really happen and how badly.
The distinction also matters for accountability. A technical team owns vulnerabilities and can be measured on patch timeframes. A security or risk function owns risk decisions, including the decision to accept a risk rather than treat it, and that decision usually needs to rest with someone senior enough to own the consequence. Treating every vulnerability as an automatic risk to escalate removes that judgement from the process.
How do frameworks like the ISM and ISO 27001 treat risk?
Both expect a documented risk assessment methodology rather than an intuitive call. The Australian Signals Directorate’s Information Security Manual expects agencies and regulated entities to apply a risk based approach to selecting controls, informed by the current threat environment ASD publishes. ISO 27001 requires a formal risk assessment and risk treatment process under clause 6, feeding the Statement of Applicability that lists which of the 93 Annex A controls apply and why.
Neither framework treats a vulnerability scan result as a finished risk assessment. Both expect the scan output to be weighed against likelihood, business impact and existing controls before it becomes a risk decision. An IRAP assessment works the same way for government systems, mapping findings against the ISM’s control set rather than reporting raw technical findings as risk on their own.
Frequently asked questions
What is the difference between a threat and a vulnerability?
A threat is a potential cause of harm, an attacker, a piece of malware, a natural event. A vulnerability is a weakness a threat can exploit, a missing patch, a misconfigured port, weak credentials. A threat needs a vulnerability to cause damage, and a vulnerability without a threat targeting it stays theoretical.
Is risk the same thing as a vulnerability?
No. A vulnerability is one input to risk. Risk is the combination of a threat, a vulnerability and the impact if that vulnerability is exploited, usually expressed as likelihood multiplied by consequence. A system can carry a known vulnerability with low risk if no realistic threat targets it or the impact is minor.
Can you have risk without a vulnerability?
Not in the technical sense used by most security frameworks. Risk in information security needs an exploitable weakness to attach to. Remove the vulnerability and that specific risk drops close to zero, even though the threat, an attacker group or a piece of malware, still exists elsewhere.
How do organisations measure risk in practice?
Most use a likelihood and consequence matrix, scoring each on a simple scale from low to extreme and plotting the combination on a heat map. Frameworks such as the Information Security Manual and ISO 27001 both expect a documented risk assessment method, not a single number treated as absolute truth.
Why does the difference between threat, vulnerability and risk matter?
Because the fix is different for each one. A threat cannot be patched, only exposure to it reduced. A vulnerability can be patched directly. Risk is what gets reported to a board, because it combines both with business impact, and it is what decides where budget goes first.
Sources:
- ISO 31000:2018, Risk management guidelines, ISO, 2018
- NIST SP 800-30 Rev 1, Guide for Conducting Risk Assessments, NIST, 2012
- OWASP Risk Rating Methodology, OWASP, accessed 2026
- Australian Signals Directorate, Information Security Manual, ASD, 2026
- ASD Annual Cyber Threat Report, Australian Signals Directorate, 2025
Last updated: 30 August, 2026
