AI Security AI安全 12h ago Updated 8h ago 更新于 8小时前 47

Android Car Malware Spreads Through Built-In Updaters for Ad Fraud, Proxy Botnet Android车载恶意软件通过内置更新程序传播,用于广告欺诈和代理僵尸网络

First documented case of malware with an infection chain specifically tailored for Android-based car head units, discovered by Kaspersky in June 2026 Malware spreads by weaponizing the legitimate built-in software update mechanism of DoFun-powered head units via the TWCore system app and an MQTT message broker Attributed with high confidence to the MoYu Group, linked to the BADBOX ad fraud and residential proxy botnet scheme The malware deploys a multi-stage downloader (JarService dropper → load 卡巴斯基发现首个专门针对Android车载信息娱乐系统的恶意软件家族,通过DoFun固件内置更新程序传播 恶意软件利用TWCore系统应用的MQTT消息代理分发多阶段下载器,实现广告欺诈和代理僵尸网络 活动与MoYu Group/BADBOX僵尸网络高度关联,攻击者持续使用版本化载荷规避检测 感染链包含信息收集、C2通信、远程命令执行等完整功能,支持9种恶意操作命令 这是车载系统供应链安全的首次重大案例,暴露了Android车载平台的安全脆弱性

72
Hot 热度
68
Quality 质量
58
Impact 影响力

Analysis 深度分析

TL;DR

  • First documented case of malware with an infection chain specifically tailored for Android-based car head units, discovered by Kaspersky in June 2026
  • Malware spreads by weaponizing the legitimate built-in software update mechanism of DoFun-powered head units via the TWCore system app and an MQTT message broker
  • Attributed with high confidence to the MoYu Group, linked to the BADBOX ad fraud and residential proxy botnet scheme
  • The malware deploys a multi-stage downloader (JarService dropper → loader → payload) that covertly operates in the background to serve ads, execute ad fraud, and build a proxy botnet
  • Supports nine commands including ad display, HTTP requests, WebView JavaScript execution, arbitrary code loading, and device information harvesting, with C2 communication every 90 minutes

Why It Matters

This represents a significant escalation in automotive cybersecurity threats, as attackers are now exploiting legitimate firmware update channels on Android-powered vehicle head units — a vector previously unreported for this device class. The convergence of ad fraud and proxy botnet operations on in-car systems highlights how the expanding attack surface of connected vehicles demands urgent, purpose-built security controls rather than generic mobile security approaches.

Technical Details

  • Infection vector: The malware exploits the TWCore system app (com.tw.core), a legitimate analytics and OTA update component that uses an MQTT message broker hosted on cardoor[.]cn to distribute APK updates. Threat actors hijacked this channel to push a malicious dropper called JarService.
  • Multi-stage payload delivery: JarService launches a loader that contacts an attacker server via HTTP POST, receives a download link for the next-stage payload (e.g., dex3.68.png), and deploys the final malware as a background user application with no UI. Seven distinct payload variants were identified (versions 3.57 through 3.68).
  • C2 communication: The implant sends a POST request to /cpc/api/task every 90 minutes, reporting device info and configuration version. The C2 server responds with either updated C2 addresses/paths or integer command identifiers (productId) mapped to serialized JSON commands stored via SharedPreferences.
  • Command set: Nine supported commands include return (SharedPreferences read), copy (clipboard manipulation), http (HTTP GET/POST), web (WebView with arbitrary JS execution), loadlib2 (download and execute arbitrary code), deeplink (browser URL opening), and traceroute (ICMP ping). The malware also downloads the "zhima" reverse proxy module, linking it to the broader BADBOX infrastructure.
  • Target ecosystem: DoFun-powered Android head units, both factory-installed and aftermarket, which feature SIM card slots for internet connectivity — making them attractive targets due to their always-on network access and growing market prevalence.

Industry Insight

  • Automotive OEMs and head unit firmware vendors must treat software update mechanisms as critical security boundaries; the abuse of legitimate OTA channels demonstrates that supply chain trust assumptions are insufficient without cryptographic verification and integrity checks on update payloads.
  • The persistence of BADBOX-affiliated actors despite law enforcement actions and botnet takedowns signals that ad fraud and proxy botnet operations are highly resilient and adaptive — security teams should expect continued diversification of infection vectors into IoT and embedded device categories.
  • The emergence of car head units as a new malware target underscores the need for automotive-specific threat modeling and security frameworks that account for the unique constraints of in-vehicle Android systems, including limited patching capabilities, long device lifecycles, and the potential for physical-world impact from compromised systems.

TL;DR

  • 卡巴斯基发现首个专门针对Android车载信息娱乐系统的恶意软件家族,通过DoFun固件内置更新程序传播
  • 恶意软件利用TWCore系统应用的MQTT消息代理分发多阶段下载器,实现广告欺诈和代理僵尸网络
  • 活动与MoYu Group/BADBOX僵尸网络高度关联,攻击者持续使用版本化载荷规避检测
  • 感染链包含信息收集、C2通信、远程命令执行等完整功能,支持9种恶意操作命令
  • 这是车载系统供应链安全的首次重大案例,暴露了Android车载平台的安全脆弱性

为什么值得看

这篇文章揭示了恶意软件攻击从传统移动设备向智能汽车领域的延伸,标志着车载信息安全进入新阶段。对于汽车制造商、Tier 1供应商和网络安全从业者而言,这是首次针对车载信息娱乐系统的专门恶意软件案例,为行业提供了宝贵的威胁情报和防御参考。

技术解析

  • 传播机制:恶意软件通过DoFun开发的Android车载信息娱乐系统固件内置更新程序传播,利用TWCore系统应用(com.tw.core)的MQTT消息代理功能,从"cardoor.cn"子域接收恶意APK文件并安装到外部缓存目录。
  • 多阶段载荷:采用JarService dropper启动loader,通过HTTP POST发送设备信息,服务器返回版本化载荷链接(如dex3.68.png),已发现7个不同版本(3.57-3.68),支持动态更新C2地址和请求路径。
  • 命令与控制:恶意软件每90分钟向C2端点发送POST请求,支持9种命令:return(读取SharedPreferences)、copy(剪贴板操作)、http(HTTP请求)、web(WebView执行JS)、loadlib2(下载执行代码)、deeplink(浏览器打开)、traceroute(ICMP ping)等。
  • 恶意功能:主要目的是广告欺诈和代理僵尸网络,可显示恶意广告、执行欺诈操作、下载额外恶意模块(如zhima反向代理),并收集设备详细信息(分辨率、型号、Wi-Fi ID、MAC地址)。
  • 供应链关联:与BADBOX僵尸网络(MoYu Group)高度关联,该组织2025年7月被Google起诉,攻击者通过IPTV应用和车载系统等多种渠道分发恶意软件,持续进行全球感染活动。

行业启示

  • 车载安全需纳入汽车安全框架:现代汽车平台亟需建立专门的恶意软件防护机制,将车载信息娱乐系统纳入ISO/SAE 21434等汽车网络安全标准范围,实施固件签名验证和更新通道安全加固。
  • 供应链安全成为关键防线:DoFun固件案例表明,第三方供应商的更新机制可能被武器化,汽车制造商需加强对Tier 1供应商的安全审计,实施端到端的供应链安全验证和持续监控。
  • 跨领域威胁情报共享至关重要:BADBOX僵尸网络同时影响移动设备、车载系统和IPTV应用,行业需要建立跨领域的威胁情报共享机制,及时识别和应对针对新兴物联网设备的恶意软件攻击。

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

Security 安全 Autonomous Driving 自动驾驶 Research 科学研究