A weakly protected application effectively prepares its own documentation for the attacker. That line is from our latest article that explores security in the AI world.
Can AI reverse engineer a mobile application? It isn’t a magic button that can turn protected binaries back into source code. But what it can do is reduce the cost of reaching the next useful result; especially if the material it’s given is clean, readable, and structured. The answer, then, isn’t to prohibit AI, or to rely on model providers to monitor and refuse malicious-looking prompts. Instead, it’s to give bad actors less to work with in the first place by protecting how the app executes, and validating signals across the whole mobile channel.
In the same article, we highlight certain conditions that a serious runtime layer has to detect. And two of them are sharpened in the latest DexProtector releases this month: new iOS runtime checks expand detection of virtualized and simulated environments, and enhanced Android protections strengthen your defense against runtime code injection.
Protection becomes more complicated when you develop an SDK rather than an app. Attackers targeting SDKs don’t even need to work with the final app; they can build a minimal harness around it instead, and instrument it at their leisure. In our latest Knowledge Hub guide, we set out why protecting an SDK is harder than protecting an app, and what SDK protection actually requires. It ends with a checklist worth holding any vendor against.
What's new with Licel's solutions?
Enhanced Protection and Precision
This month’s DexProtector releases deliver stronger protection, and more precise security insights across both iOS and Android.
On iOS, new runtime checks expand detection of virtualized and simulated environments, helping applications identify potentially untrusted execution conditions more effectively. Network security mechanisms have also been refined, with optimized Public Key Pinning and Certificate Transparency checks for hostnames resolving to multiple IPv4 and IPv6 addresses.
Alice reporting now includes additional component information, making it easier to identify and filter events originating from extension processes.
On Android, enhanced protections provide stronger defense against runtime code injection attacks. The latest updates also improve compatibility for applications targeting Android 17 and those using custom AppComponentFactory implementations.
SDK Protection: How to Secure Your Most Sensitive Asset in its Most Exposed Form
The difference between an exposed mobile application and an exposed SDK is often the blast radius. One published bypass affects every customer who has embedded it.
Because an SDK is incorporated into a host application, as an SDK developer you have to accept a certain loss of control, because your code is packed up and shipped inside an application that you don’t know.
There’s the rub, and the reason why SDK protection is more complicated than application protection. This lack of control and visibility also means that integrating robust protection is absolutely vital for SDK developers. Especially when you consider that these days, SDKs often contain the very things that bad actors want to attack; things like payment logic and cryptographic operations.
We’ve written a Knowledge Hub guide that explains what makes SDK protection different, how DexProtector confronts the challenges, and a checklist of six non-negotiable requirements you should demand when evaluating Android or iOS protection tools and obfuscators.
Is AI an Angel or a Demon? Finding the Right Security Balance
Stories about AI and its abilities can sometimes be a little sensationalist and exaggerated. On its own, AI can’t completely dissect your application and understand every instruction perfectly. But what it can do, and is already doing, is help attackers to reach their goals more quickly, efficiently, and cheaply. And that’s a lot.
The modern-day attacker, assisted by AI, is more brazen than ever. They won’t respect the boundaries between your mobile development team, fraud team, API team, and security operations center. So, your protection architecture shouldn’t either.
In our latest long read, we take a deep dive into security in the post-AI era, examining whether it’s accurate or helpful to categorize AI as either good or bad. We also explore whether AI can reverse engineer an Android or iOS application, and take a look at context bombs and other innovative techniques that are being used to balance the scales.
Thanks for reading this edition of the Licel Layers Bulletin. We'll be back next month with more product improvement updates and threat intelligence insights.
