STEPN Upgrades its Software with AI-Powered Anti-Cheating Model
The popular move-to-earn app STEPN has released an AI-powered update called “STEPN Model for Anti-Cheating” or SMAC. This update is meant to make the M2E app more fair and transparent.
Along with the new update, STEPN will focus on additional resources to maintain the platform’s integrity. The AI-powered system leverages deep data analytics and real-time tracking to achieve its objectives.
STEPN has invested in technologies driven by AI that can monitor the activities on the M2E app and identify potential instances of cheating. The app’s artificial intelligence technology is built on "self-learning algorithms" that employ a vast quantity of user-related data. This includes motion sensors, GPS tracking, and personal health information. The objective is to evaluate all of this data and identify any irregularities. Even contextual and collective outliers are incorporated into the analysis to reduce false positives.
The system particularly monitors running and walking data and cross-references it with other data points to find suspicious patterns. Once a user has been flagged, all STEPN awards are confiscated and their account is deleted. However, after SMAC released the upgrade, it mistakenly identified many users as bots.
In a blog post discussing the current upgrade, the company wrote:
"It is natural that when an app gets popular, many are tempted to abuse or exploit the system, causing developers to have their hands full with detecting and banning those who cheat. But how does the system determine a genuine anomaly from one that is fraudulent in nature? Simply put, by fixing a specified threshold value for reconstruction errors, and then cross referencing it across multiple autoencoder data points."
Additionally, the SMAC system will try to eliminate cheating from STEPN’s NFT marketplace by disabling buy/sell scripts and deleting any mining bots.
The M2E application will also generate a "Turing Score" which is basically an integrity ranking system. Every user will receive a Turing Score of 100/100 at the beginning of each walking or running session, and this score might increase or decrease according to how they engage with the STEPN app.