Rotten Tomatoes Introduces AI-Powered Review Filter to Combat Viral Bot Score Manipulation
SAN FRANCISCO, CA - In a decisive move, the review aggregation website Rotten Tomatoes has announced the implementation of an artificial intelligence system designed to detect and filter out fraudulent audience reviews. This initiative, according to a company spokesperson, aims to counteract coordinated bot-driven campaigns that have artificially inflated or deflated scores for specific films, a trend that has significantly undermined the platform's credibility in recent months.
The new algorithm, which analyzes over 50 data points including writing patterns, account history, and voting speed, will retroactively audit reviews for recent high-profile releases. The company reports that its beta testing phase successfully identified and removed over 15,000 fraudulent reviews for a single blockbuster title. This development comes as the film industry grapples with growing concerns over the integrity of online ratings, with some estimates suggesting that up to 25% of all audience scores on the platform may be inauthentic.
A technical analysis reveals that the AI filter is particularly effective at detecting language patterns common in promotional posts or overly polarized criticism. The system is currently integrated into the platform's "Verified" audience score feature, which requires proof of ticket purchase. However, industry experts note that the core challenge remains: preventing coordinated manipulation of the site's main "Audience Score," which does not require ticket verification.
The announcement has prompted mixed reactions. Filmmakers and major studios have expressed cautious optimism, with one prominent director stating, "Rotten Tomatoes must be defended as a bastion of honest critique, not a battlefield for fan armies." Conversely, some user advocacy groups have raised concerns about potential censorship, urging the company to maintain transparency in its moderation process. The company has confirmed that all AI-flagged reviews will be subject to human review before removal, as part of a protocol to minimize errors.