(2014) Poster: Quasi-ID: In fact, I am a human.
Abstract
CAPTCHAs (Completely Automated Public Turing test to tell Computers and Humans Apart) are the dominantly used turing tests to protect websites against bots that are impersonating human users to gain access to various types of services. The test is designed in a way to be very difficult for robotic programs, but comfortably easy for humans. As artificial intelligence research thrives towards the biggest challenge of the field — simulating the work of a human brain — the complexity of CAPTCHA tests increases, making it more and more difficult for humans to answer the tests. The problem gets even bigger, with the latest research reports in fact indicating that CAPTCHAs are broken. We present Quasi-ID: a novel approach for determining whether or not a user is a human in a scalable and privacy-preserving manner. Our system utilizes smart devices as ubiquitous input devices for invoking a physical interaction with the users. Such an interaction between the user and his smart device can prove that the user is actually a human. Support for Quasi-ID can be deployed today along with the current CAPTCHA solutions. It does not add additional burden to the web service and requires a non-persistent communication with the Quasi-ID service provider.
| Item Type: | Conference or Workshop Item (A Paper) (Poster) |
|---|---|
| Additional Information: | pub_id: 99 Bibtex: Simeonovski:QuasiID URL date: None |
| Uncontrolled Keywords: | authentication,group:infsec,privacy,two-step verification |
| Divisions: | Michael Backes (InfSec) |
| Conference: | CCS ACM Conference on Computer and Communications Security |
| Depositing User: | Sebastian Weisgerber |
| Date Deposited: | 26 Jul 2017 10:31 |
| Last Modified: | 18 Jul 2019 12:12 |
| Primary Research Area: | NRA5: Empirical & Behavioral Security |
| URI: | https://publications.cispa.saarland/id/eprint/721 |
Actions
Actions (login required)
 |
View Item |