ZKSENSE: A Friction-less Privacy-Preserving Human Attestation Mechanism for Mobile Devices
Bibliographic record
Abstract
Abstract Recent studies show that 20.4% of the internet traffic originates from automated agents. To identify and block such ill-intentioned traffic, mechanisms that verify the humanness of the user are widely deployed, with CAPTCHAs being the most popular. Traditional CAPTCHAs require extra user effort (e.g., solving mathematical puzzles), which can severely downgrade the end-user’s experience, especially on mobile, and provide sporadic humanness verification of questionable accuracy. More recent solutions like Google’s reCAPTCHA v3, leverage user data, thus raising significant privacy concerns. To address these issues, we present zkSENSE: the first zero-knowledge proof-based humanness attestation system for mobile devices. zkSENSE moves the human attestation to the edge: onto the user’s very own device, where humanness of the user is assessed in a privacy-preserving and seamless manner. zkSENSE achieves this by classifying motion sensor outputs of the mobile device, based on a model trained by using both publicly available sensor data and data collected from a small group of volunteers. To ensure the integrity of the process, the classification result is enclosed in a zero-knowledge proof of humanness that can be safely shared with a remote server. We implement zkSENSE as an Android service to demonstrate its effectiveness and practicality. In our evaluation, we show that zkSENSE successfully verifies the humanness of a user across a variety of attacking scenarios and demonstrate 92% accuracy. On a two years old Samsung S9, zkSENSE’s attestation takes around 3 seconds (when visual CAPTCHAs need 9.8 seconds) and consumes a negligible amount of battery.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.013 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.005 |
| Open science | 0.003 | 0.008 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 0.003 |
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; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".