Skip to main content
COSMICBYTEZLABS
NewsSecurityHOWTOsToolsTraining
StudyProjectsNewsletterHire MeAbout
Subscribe

Press Enter to search or Esc to close

News
Security
HOWTOs
Tools
Training
Study
Projects
Newsletter
Hire Me
About
RSS Feed
Reading List
Subscribe

Stay in the Loop

Get the latest security alerts, tutorials, and tech insights delivered to your inbox.

Subscribe NowFree forever. No spam.
COSMICBYTEZLABS

Your trusted source for IT intelligence, cybersecurity insights, and hands-on technical guides.

2891+ Articles
167+ Guides

CONTENT

  • Latest News
  • Security Alerts
  • HOWTOs
  • Checklists
  • Projects
  • Exam Prep

RESOURCES

  • Search
  • Browse Tags
  • Newsletter Archive
  • Reading List
  • RSS Feed

COMPANY

  • About Us
  • Contact
  • Privacy Policy
  • Terms of Service

© 2026 CosmicBytez Labs. All rights reserved.

System Status: Operational
  1. Home
  2. News
  3. New RatHat Android Malware Uses AI to Automate Device Control
New RatHat Android Malware Uses AI to Automate Device Control
NEWS

New RatHat Android Malware Uses AI to Automate Device Control

Zimperium found RatHat, an Android RAT whose AI subsystem reads accessibility trees to auto-navigate banking apps and steal credentials.

Dylan H.

News Desk

September 17, 2026
3 min read

Zimperium's zLabs research team has discovered a new Android remote-access trojan called RatHat that hands its operators an unusual advantage over previous banking malware: an AI subsystem that reads a device's screen and decides how to navigate it, rather than following a fixed, scripted sequence of taps and swipes.

Distribution

RatHat doesn't come through the Google Play Store. Zimperium found it spreading via malvertising campaigns, SMS messages, and phishing sites that push direct APK downloads, all pretexts designed to get a victim to sideload the app outside Google's normal vetting.

The AI-Powered Automation

Once installed, RatHat abuses Android's Accessibility permissions — the same API legitimate screen readers and automation tools rely on — to read the device's UI. Its distinguishing feature converts the Android accessibility tree into XML, identifies the on-screen coordinates and text of interface elements, and then generates navigation commands like "SCROLL_DOWN" to move through an app. Zimperium notes this approach is "more adaptable and harder for security software to detect than traditional, scripted automation," because the malware is reasoning about whatever screen it's actually looking at rather than replaying a hardcoded script that breaks the moment an app's layout changes.

Researchers also found Chinese-language LLM prompts embedded in the malware's code, pointing toward a Chinese-speaking threat actor, though Zimperium has not attributed RatHat to a specific named group.

Capabilities

Beyond AI-driven navigation, RatHat's accessibility abuse enables a fairly complete device-takeover toolkit:

  • Enabling Developer Options and Wireless Debugging to open up local shell access
  • Installing a Go-based agent that runs with ADB shell privileges
  • Displaying HTML overlays on top of banking and cryptocurrency apps to harvest login credentials
  • Intercepting SMS messages, push notifications, and one-time passwords
  • Recording keystrokes and capturing lock-screen PINs
  • Reading URLs directly out of the browser's address bar

Technical Indicators

The malware ships with two native agents (liblocal-service.so and libmedia_codec.so), an unusually bloated 61MB Android manifest, and invalid DEX pseudo-instructions apparently inserted to trip up static-analysis tooling. For remote access, it establishes a persistent tunnel using FRP, an open-source reverse-proxy tool commonly repurposed by malware operators for exactly this kind of covert command-and-control channel.

Why It Matters

RatHat is a preview of where mobile banking trojans are heading: instead of a fragile, hand-coded automation script that a defender can fingerprint by its exact tap sequence, an AI-driven agent adapts to whatever UI it encounters, including redesigns and app updates that would normally break scripted malware outright. Combined with accessibility-permission abuse — already one of the hardest classes of Android malware to fully block without disabling legitimate assistive technology — this raises the bar for behavioral detection meaningfully.

Protection Tips

  • Never sideload APKs from ad links, SMS messages, or unfamiliar sites — install apps only through the Google Play Store.
  • Be suspicious of any app requesting Accessibility permissions unless it's a screen reader, automation tool, or another app you specifically installed for that purpose.
  • Watch for unexplained Developer Options or Wireless Debugging being enabled — that's a strong indicator of this exact abuse pattern.
  • Use hardware or app-based two-factor authentication rather than SMS OTPs where possible, since RatHat is built to intercept SMS-delivered codes.

Sources

  • BleepingComputer — New RatHat Android malware uses AI to automate device control
#Android#Malware#Mobile Security#AI#Banking Trojan

Related Articles

ToxicPanda 2.0 Android Banking Trojan Abuses VPN Permissions to Neutralize Google Play Protect

ToxicPanda 2.0 targets 349 financial apps across 16 countries, using VPN hijacking and ADB abuse to bypass Android security.

4 min read

Android 17 Blocks Non-Accessibility Apps from Accessibility

Google is testing a new Android Advanced Protection Mode enforcement in Android 17 Beta 2 that automatically strips non-accessibility apps of their...

6 min read

SCMBANKER Malware Uses ClickFix Lures to Target Mexican Banking Users

A new banking fraud campaign tracked as REF6045 is deploying SCMBANKER malware through fake CAPTCHA ClickFix lures to steal credentials from customers of...

3 min read
Back to all News