Across Inspectiv's bug bounty programs, 29% of all submitted issues fall under Broken Access Control, more than any other category, and it's also the top spot on OWASP's Top 10 list. We've written before about why authorization breaks down so often. What’s more compelling than that headline number is what’s underneath it: the issue patterns making up that 29% differ sharply in frequency, severity, and how far each one reaches.
Inside Inspectiv's own submission data, the three most common Broken Access Control issue patterns split out like this:
54% are privilege escalation, the most frequently discovered pattern.
18% are insecure direct object reference (IDOR), which tends to affect the widest range of users at once, especially in multi-tenant products.
9% are broken authentication, the rarest of the three, but consistently the most severe when it does show up.
Knowing which one you're likely to have changes how you test for it and how urgently it needs fixing.
A user performs an action their role was never scoped for, because the backend never checks whether they're allowed to. A common example: a low-privilege account adds itself, or another user, as an internal admin, simply because the API endpoint that handles that action doesn't verify the caller's role. This is the pattern Inspectiv's researchers find most often, largely because permission checks tend to get added secondhand as a product grows, rather than enforced consistently from a single source of truth.
A user accesses or modifies a data object they were never granted access to, often just by changing an ID in a URL or API call. In a multi-tenant SaaS product, this can mean one customer viewing or editing another customer's records. IDOR shows up less often than privilege escalation in Inspectiv's data, but when it does, it tends to expose the largest number of affected users at once, since the same flaw usually applies across an entire tenant boundary, not just one account.
An endpoint, panel, or application is reachable with no authentication check at all, such as an internal admin panel that's exposed without requiring a login. This is the least common of the three patterns in Inspectiv's data, but it's consistently the highest severity, since there's no access boundary to get past in the first place.
A few things are making broken access control more persistent, not less:
AI-assisted development means code ships faster than ever, and by default it tends to be insecure until someone specifically checks it.
Security review often isn't built into the software development lifecycle from the start, so access control logic gets bolted on rather than designed in.
Standard penetration testing is limited in scope and time, and frequently skips authenticated, multi-role testing, which is exactly the kind of testing that surfaces privilege escalation and IDOR.
We've covered the volume side of this shift in more detail in our post on AI and bug bounty hunting; the same forces that produce more low-effort reports also produce more shippable, unreviewed code.
No testing: no detection. Issues sit in production until a researcher, or an attacker, finds them.
Pentesting only: partial detection. Point-in-time and scope-limited, and authenticated multi-role testing often gets cut for time.
Pentesting plus continuous bug bounty testing: the most effective combination Inspectiv sees, since ongoing researcher attention catches what a single scheduled test misses between releases.
As industry voice Benson Wu put it: “By building security in, we can change the game of cyber attacks rather than just raise the bar.”
Build a complete user role access matrix as a base building block before writing authorization logic, not after the fact.
If you're running automated scans, test with at least two different user contexts. A single-context scan structurally can't catch privilege escalation or cross-tenant IDOR, since there's no second account to compare against.
Default to deny during the design phase, and grant access explicitly, rather than assuming something is safe unless it's specifically blocked.
Inspectiv's platform combines bug bounty and penetration testing so access control gets tested continuously, not just once a year, and every submission is validated by Inspectiv's triage team before it reaches a customer. That combination is built to catch the most common Broken Access Control issue patterns, the frequent privilege escalation issues, the wide-reaching IDORs, and the rare but severe authentication gaps, along with whatever else turns up outside those three, without a company having to wait for its next scheduled pentest to find out what it actually has.
If you want to see where your own access control gaps fall across these issue patterns, reach out to get a demo.