AI-Written Apps Are Shipping With Hundreds of Security Holes, Study Finds
A new analysis counted 434 exploitable flaws across apps built with AI coding tools. The findings raise hard questions about who is responsible when software writes itself.
Key points
- Researchers found 434 exploitable security flaws across a set of applications built using AI coding assistants.
- The most common problems included denial-of-service vulnerabilities (where an attacker can crash or overload an app), broken authorisation controls (where users can access things they should not be allowed to see), and secrets exposure (where passwords or private keys are accidentally left visible in the code).
- The findings suggest that AI coding tools reproduce insecure patterns at scale, not just occasionally.
- No single vendor's tool was named as uniquely responsible; the problem appears to span the category.
Software is getting built faster than ever. AI coding assistants, sometimes called "vibe coding" tools because developers describe what they want in plain language and the AI writes the actual code, have made it possible for almost anyone to ship a working application in an afternoon. That speed comes with a catch.
A new analysis, covered by SecurityWeek, examined apps produced with these AI tools and found 434 flaws that a real attacker could actually use. Not theoretical weaknesses. Practical ones.
The three biggest categories tell a clear story. First, denial-of-service weaknesses, meaning an attacker can send a flood of requests that overwhelms the app and knocks it offline for everyone else. Second, broken authorisation, where the app fails to properly check who is allowed to do what, so one user might read another user's private data or perform actions reserved for administrators. Third, secrets exposure, where the AI accidentally bakes passwords, private encryption keys, or database credentials directly into the code where they can be found and stolen.
Why should ordinary users care about this?
If an app you use was built with an AI coding tool and that app holds your personal details, a payment method, or health records, these flaws create real doors for criminals to walk through. Broken authorisation alone could let a stranger read your private account information without ever guessing your password.
The problem is structural. AI models learn from enormous collections of existing code, and a lot of that existing code is insecure. The model reproduces those patterns confidently and at speed. A developer who is not a security expert, which describes many people now building apps with these tools, may have no way to spot the problem before they hit publish.
From a threat-intelligence perspective, this matters beyond individual apps. At medium confidence, I'd note that organised criminal groups and state-linked hackers alike scan for exactly the categories of flaw listed here. Secrets exposure, in particular, is a known initial-access technique: automated tools scrape publicly visible code repositories looking for leaked credentials, then use them to break into backend systems.
If you use apps built by small teams or solo developers, a few practical steps make sense. Watch for any unexpected emails asking you to reset passwords you did not change. Check whether the app offers two-factor authentication, which requires a second code in addition to your password, and turn it on. If the service holds financial data, review your statements regularly.
Building fast is not the problem. Building fast without a security check at the end is.
