Sovra Recon·Glossary · Access Control

Broken Access Control: Definition and Examples

By Solan Barlet, co-founder of Sovra Recon
Updated October 29, 2026

In short

Broken Access Control refers to any vulnerability that allows a user to act beyond their authorized permissions: reading another customer’s data, modifying an account that does not belong to them, or accessing administrative functions. This is the No. 1 category in the OWASP Top 10, both in the 2021 and 2025 editions. It is difficult to detect automatically because only the business context determines who is authorized to view what.

Real-world examples

  • IDOR: Changing /invoices/1042 to /invoices/1043 displays another customer’s invoice.
  • Vertical escalation: A standard user directly calls an administrative endpoint that the interface does not display to them.
  • Horizontal escalation: An employee of one client company accesses the data of another client company using the same SaaS application.
  • Method-based bypass: an action that is denied via a GET request is accepted via a POST or PUT request.
  • Token Manipulation: A role stored in a cookie or a mal-signed JWT is modified on the client side.

How to Protect Against It

  • Deny by default: all access must be explicitly authorized.
  • Verify permissions on the server side, for every request, on every object—never just in the user interface.
  • Centralize authorization logic rather than duplicating it in every route.
  • Test using multiple accounts at the same level and at different levels.

Frequently asked questions

Why do scanners have trouble detecting Broken Access Control?

This is because a “200 OK” response does not indicate whether the user was authorized to view the data. One must understand the application and compare what different accounts see—which is what a penetration tester or an AI agent does.

Sources

Find out where an attacker would get in.

Our AI agents actually attack your scope, fully automated, and hand you the proof of every vulnerability.