Fake Reviews

Artificial ratings inflate public discourse and endorse incentivised goods and services

Fraudsters generate bogus ratings for goods and services by leveraging a variety of automation tools to sway commercial influence and public opinion. The anonymity, credibility, and independence of online reviews make lucrative targets for abuse that seeks to game the system. Human-assisted attack vectors are also exploited by fraudsters who use phone farms, Amazon Mechanical Turk (MTurk), and digital sweatshops.

Fake Reviews are are also known as Bot Likes, Fake Ratings, Fraudulent Reviews, and Fake Customer Testimonials.

Arkose Labs challenge Fake Reviews and model inauthentic requests

Fake Reviews are a common nuisance for customers, and enterprises continue to invest significant resources towards protecting the integrity of their review applications. Arkose Labs intercept Fake Reviews with Enforcement, a challenge–response mechanism that inhibits automation by verifying that each request is being made under authentic conditions. Authentication results are subsequently passed back to Telemetry, and trained to accurately model authentic/inauthentic results with machine learning. This validates how the system makes decisions and incrementally limits challenges to inauthentic requests only.

Fake Reviews deceive customers and lead to a loss of user trust

The rise of Single Request Attacks have facilitated a disguise for Fake Reviews that can be generated to look authentic at scale. These requests imitate legitimate ratings by obfuscating IP addresses, consuming dynamic fingerprints, using headless browsers, and executing JavaScript as expected. User trust relies on the enterprise being able to classify deceptive endorsements with definitive accuracy.


Customers value the insights and experiences shared online by fellow shoppers

Find out how Arkose Labs can protect the integrity of your rating web application and stop fraudsters