I performed a fixed analysis of DeepSeek, securityholes.science a Chinese LLM chatbot, utilizing variation 1.8.0 from the Google Play Store. The objective was to determine potential security and personal privacy issues.
I have actually composed about DeepSeek previously here.
Additional security and personal privacy issues about DeepSeek have been raised.
See likewise this analysis by NowSecure of the iPhone variation of DeepSeek
The findings detailed in this report are based simply on fixed analysis. This means that while the code exists within the app, there is no definitive proof that all of it is carried out in practice. Nonetheless, the existence of such code warrants scrutiny, particularly offered the growing concerns around data personal privacy, security, the prospective misuse of AI-driven applications, and cyber-espionage characteristics in between global powers.
Key Findings
Suspicious Data Handling & Exfiltration
- Hardcoded URLs direct information to external servers, raising issues about user activity monitoring, such as to ByteDance "volce.com" endpoints. NowSecure determines these in the iPhone app the other day too.
- Bespoke file encryption and data obfuscation approaches exist, with signs that they could be used to exfiltrate user details.
- The app contains hard-coded public secrets, rather than relying on the user device's chain of trust.
- UI interaction tracking records detailed user behavior without clear consent.
- WebView control exists, which might permit the app to gain access to personal external browser information when links are opened. More details about WebView manipulations is here
Device Fingerprinting & Tracking
A significant portion of the analyzed code to focus on event device-specific details, which can be utilized for tracking and fingerprinting.
- The app collects various special device identifiers, including UDID, Android ID, IMEI, wiki.vst.hs-furtwangen.de IMSI, and provider details. - System residential or commercial properties, set up packages, and root detection mechanisms suggest possible anti-tampering measures. E.g. probes for the presence of Magisk, a tool that privacy advocates and security scientists use to root their Android devices.
- Geolocation and network profiling are present, suggesting potential tracking abilities and enabling or disabling of fingerprinting routines by area.
- Hardcoded gadget model lists suggest the application might behave differently depending upon the found hardware.
- Multiple vendor-specific services are utilized to draw out extra gadget details. E.g. if it can not figure out the device through standard Android SIM lookup (since approval was not granted), it attempts producer specific extensions to access the very same details.
Potential Malware-Like Behavior
While no definitive conclusions can be drawn without dynamic analysis, numerous observed behaviors align with known spyware and malware patterns:
- The app utilizes reflection and UI overlays, which might help with unauthorized screen capture or phishing attacks. - SIM card details, serial numbers, and other device-specific information are aggregated for unknown purposes.
- The app executes country-based gain access to constraints and "risk-device" detection, suggesting possible security mechanisms.
- The app executes calls to load Dex modules, where additional code is filled from files with a.so extension at runtime.
- The.so files themselves turn around and make extra calls to dlopen(), which can be used to pack additional.so files. This facility is not usually examined by Google Play Protect and other static analysis services.
- The.so files can be executed in native code, such as C++. Making use of native code adds a layer of complexity to the analysis procedure and obscures the full extent of the app's capabilities. Moreover, native code can be leveraged to more quickly escalate privileges, library.kemu.ac.ke potentially exploiting vulnerabilities within the operating system or device hardware.
Remarks
While data collection prevails in contemporary applications for debugging and enhancing user experience, aggressive fingerprinting raises significant privacy concerns. The DeepSeek app needs users to log in with a valid email, which must already provide enough authentication. There is no valid factor for the app to aggressively gather and transmit distinct device identifiers, IMEI numbers, SIM card details, and other non-resettable system properties.
The degree of tracking observed here exceeds normal analytics practices, possibly making it possible for persistent user tracking and re-identification throughout gadgets. These habits, integrated with obfuscation techniques and network interaction with third-party tracking services, necessitate a higher level of analysis from security scientists and users alike.
The employment of runtime code packing as well as the bundling of native code suggests that the app might enable the implementation and execution of unreviewed, remotely delivered code. This is a serious potential attack vector. No proof in this report exists that from another location deployed code execution is being done, just that the center for this appears present.
Additionally, the app's method to finding rooted devices appears extreme for an AI chatbot. Root detection is typically warranted in DRM-protected streaming services, where security and content security are important, or asteroidsathome.net in competitive computer game to prevent unfaithful. However, there is no clear reasoning for such strict steps in an application of this nature, raising further concerns about its intent.
Users and companies thinking about setting up DeepSeek ought to understand these potential dangers. If this application is being used within a business or federal government environment, extra vetting and security controls ought to be imposed before allowing its deployment on managed devices.
Disclaimer: The analysis presented in this report is based upon static code evaluation and does not suggest that all discovered functions are actively utilized. Further investigation is needed for definitive conclusions.