Bibliographic record
Abstract
As privacy features in Android operating system improve, privacy-invasive apps may gradually shift their focus to non-standard and covert channels for leaking private user/device information. Such leaks also remain largely undetected by state-of-the-art privacy analysis tools, which are very effective in uncovering privacy exposures via regular HTTP and HTTPS channels. In this study, we design and implement, ThirdEye, to significantly extend the visibility of current privacy analysis tools, in terms of the exposures that happen across various non-standard and covert channels, i.e., via any protocol over TCP/UDP (beyond HTTP/S), and using multi-layer custom encryption over HTTP/S and non-HTTP protocols. Besides network exposures, we also consider covert channels via storage media that also leverage custom encryption layers. Using ThirdEye, we analyzed 12,598 top-apps in various categories from Androidrank, and found that 2887/12,598 (22.92%) apps used custom encryption/decryption for network transmission and storing content in shared device storage, and 2465/2887 (85.38%) of those apps sent device information (e.g., advertising ID, list of installed apps) over the network that can fingerprint users. Besides, 299 apps transmitted insecure encrypted content over HTTP/non-HTTP protocols; 22 apps that used authentication tokens over HTTPS, happen to expose them over insecure (albeit custom encrypted) HTTP/non-HTTP channels. We found non-standard and covert channels with multiple levels of obfuscation (e.g., encrypted data over HTTPS, encryption at nested levels), and the use of vulnerable keys and cryptographic algorithms. Our findings can provide valuable insights into the evolving field of non-standard and covert channels, and help spur new countermeasures against such privacy leakage and security issues.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.006 | 0.010 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".