An AI found 24 vulnerabilities in Android apps: what should be done?

An AI combed through Android apps and found 24 flaws. Among them, a maps app downloaded more than ten million times, which could let any other app on your phone track your routes. And the Wikipedia app, where a simple malicious link could make your account change hands.

It was the GitHub Security Lab that reported it on September 28, GitHub's team of vulnerability hunters, the big site where developers from all over the world store their code. And the most interesting thing isn't the number. It's that they are giving their recipe to everyone.

Your GPS guides you, and someone else takes notes

How your maps app could tell someone where you're going

The app is called OsmAnd. It's a free navigation app, much loved by hikers and cyclists because it works even without a network.

To understand the flaw, you need to know how a map is displayed on your phone. It doesn't arrive all at once : it's split into little squares, “tiles”, which the app fetches as you move. You drive toward Namur, the app asks for the squares of the road toward Namur. Makes sense.

Except OsmAnd had left a door ajar. A door designed to import settings, and supposed to open only for the app itself. In practice, any other app installed on the phone could slip settings through it on the sly, without any particular permission and without you seeing a thing. For example this one : “from now on, fetch your tiles from MY server”.

Tell me which bits of the map you ask for, and I'll tell you where you are

And then, it's foolproof : if it's the pirate's server delivering the pieces of the map to you, it knows which ones you ask for, therefore where you are, and which way you go. Your route draws itself in its logs. According to GitHub, the same flaw could also reveal your routes, including starting and ending points. Meanwhile, you keep driving along peacefully with a map that displays perfectly. That's what gives me the creeps : there's nothing to see.

Wikipedia : a malicious link, and your account changes hands

Second detailed example, the Wikipedia app for Android. It accepts “deep links”, those links that open a page directly in the app instead of the browser. The problem : it checked the destination address badly. By chaining together two verification errors, a pirate could make it load a page of its own, retrieve the little files that keep you logged in (the “cookies”) and take control of your Wikipedia account, and of the sister sites run by the same foundation.

Okay, a stolen Wikipedia account isn't your bank account. But think about the people who edit articles read by millions of people. A respected contributor's account in the wrong hands is an open door to disinformation.

The AI finds, the human sorts

Now, the part that interests me most, as a developer. The researcher, Kevin Stubbings, didn't paste an app's code into an AI and ask “find me some flaws”. That gives you anything. He broke the work down into small steps, like a checklist : first spot every door through which another app can get in, then check each door, one by one, against a list of known traps on Android. GitHub calls these recipes “taskflows”, and they are freely available.

The conclusion he draws from it is refreshingly honest. AI is very good at finding things. It's bad at judging how serious they are. He says it often flagged flaws that assumed a situation “almost impossible to encounter in real life”. On the other hand, when you asked it to create proof that the flaw existed, the little program that triggers it, you “almost had nothing to tweak”.

The machine brings back everything that shines, the expert sorts the gold from the beer caps

It matters, for a simple reason: OsmAnd isn't a multinational corporation. It's an open source project, run by a small team, like thousands of apps you use without knowing it. Teams like these never pay several thousand euros for a security audit. A workflow that runs in one or two hours on a medium-sized project, with a Copilot subscription, GitHub's programming assistant, is an audit within anyone's reach. I was already saying it in August: these tools are cleaning up years of old code, and that's excellent news.

The downside is obvious, and GitHub isn't hiding it: a public workflow also serves those looking for vulnerabilities to exploit. The race is on from both sides. Might as well let the good guys use it first!

In practice, on your phone

Don't worry, for the OsmAnd vulnerability, a malicious app first had to be installed on your phone. Nobody could track you from the other side of the world without that accomplice in your pocket. GitHub doesn't give the fixed version numbers in its post, but its team has a rule of notifying developers before publishing anything, and giving them time to fix it.

So, three habits that apply to these two apps and all the others:

• turn on automatic updates in the Play Store. That's how fixes get to you without you having to think about it ;
• install your apps from the Play Store, not from a file found on a forum or sent by a stranger ;
• tidy things up from time to time. The flashlight installed in 2019 and never opened again is exactly the kind of app nobody monitors anymore.

Three actions, five minutes, and one fewer potential accomplice in your pocket

What I take away from this is that we're reaching a point where a machine rereads the code of free apps that nobody had time to reread. Twenty-four vulnerabilities at once, in apps we thought were harmless. Come on, go tidy up your phone tonight: I bet you'll find a flashlight in there!

Sources

• GitHub Blog, September 28, 2026: how we found 24 Android vulnerabilities with our open source AI security agent
• Help Net Security, September 29, 2026: GitHub's AI agent found 24 vulnerabilities in Android apps
• GitHub Security Lab: published security advisories

Article written with the help of Claude Code, reread and corrected by me.