Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
The principle of least privilege means every user, application, and system gets only the minimum access needed to do its job, for only as long as it is needed.

What Is the Principle of Least Privilege?
The principle of least privilege (PoLP) states that every identity—a person, a service account, an application, or a cloud workload—should hold only the permissions it needs to perform its task, and nothing more. When the task ends, the extra access should end too.
The idea is not new. Jerome Saltzer and Michael Schroeder listed it among their design principles for protecting computer systems in 1975. What has changed is the scale: a mid-sized organisation today manages thousands of human and machine identities across on-premises directories, SaaS platforms, and several cloud providers, and every unnecessary permission is a potential route for an attacker.
Least privilege does not prevent the initial compromise. It limits what happens next. If a phished user can read only their own team’s files, the attacker gets a small haul. If the same user is a local administrator with access to every share, the attacker can dump credentials, disable security tools, and begin lateral movement within minutes.
The 2019 Capital One breach illustrates the point in the cloud. An attacker exploited a server-side request forgery flaw in a misconfigured web application firewall to obtain temporary credentials for the role attached to it. That role could list and read far more Amazon S3 buckets than the firewall ever needed, and data on more than 100 million people was exposed. Tighter permissions on a single role would have contained the damage.
Few organisations grant excessive access on purpose. It accumulates:
"Action": "*" to cloud roles to make a deployment work, intending to tighten them later.Over time this creates a gap between the access people have and the access they actually use. Cloud providers’ own analysis tools routinely show that most granted permissions are never exercised.
Day-to-day accounts should not be administrators. Give IT staff separate admin accounts, and replace permanent privileged roles with just-in-time elevation: Microsoft Entra Privileged Identity Management and most PAM tools let users activate a role for a few hours with approval and logging.
Role-based access control works only if roles reflect what people actually do. Use access logs and tools such as AWS IAM Access Analyzer or Google Cloud’s IAM Recommender to identify unused permissions, then remove them in stages rather than all at once.
Service accounts often have broad rights, never sign in interactively, and are excluded from MFA. Inventory them, assign an owner to each, use group managed service accounts or workload identities where possible, and rotate secrets automatically.
Removing local admin rights from workstations blocks a large share of commodity malware, which often needs those rights to install persistence. Application allowlisting, sudo rules on Linux, and container security contexts that drop root privileges extend the principle to software itself.
Run regular access reviews, prioritising privileged and sensitive access. Reviews work best when owners see what each permission does and when it was last used, rather than a list of group names.
Cyber Essentials requires that users have only the access they need and that administrator accounts are used solely for administrative tasks, not for email or web browsing. ISO 27001 Annex A includes controls on access rights and privileged access. The NCSC treats least privilege as a foundation of zero trust architecture, and UK GDPR’s data minimisation principle applies the same thinking to personal data: if someone does not need it, they should not see it.
Useful metrics include the number of standing domain or global administrators, the percentage of privileged access granted just in time, the count of service accounts without an owner, and the average time to revoke access for leavers. These figures make progress visible to leadership and highlight where privilege escalation paths still exist.
Excessive cloud permissions are among the most common findings in our cloud security services engagements, where we map risky roles and recommend tighter policies.
It can if it is implemented clumsily, for example by removing rights without warning or making every request wait days for approval. Done well, it barely affects daily work. Self-service requests, pre-approved roles for common tasks, and just-in-time elevation that takes a minute to activate keep people productive while ensuring powerful access is no longer permanently available to attackers.
Just-in-time access grants elevated permissions only when they are needed and removes them automatically after a set period, often one to eight hours. A user requests a role, provides a reason, and may need approval or MFA before activation. Every activation is logged. This replaces permanent administrator accounts, which are attractive targets because they work around the clock.
Zero trust assumes no user or device is trusted by default and that every request must be verified. Least privilege defines what a verified identity is allowed to do once access is granted. The two work together: zero trust checks who you are and whether your device is healthy, while least privilege ensures that, even if verification is fooled, the damage remains limited.
Start with the accounts that could cause the most damage. Count your domain administrators, global cloud administrators, and service accounts with privileged rights, then reduce them. Remove local administrator rights from standard workstations and deploy unique local admin passwords. These steps deliver the greatest risk reduction quickly and build momentum for broader role clean-up later.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.