AI in Casinos: How Artificial Intelligence Spots Dealer Errors and Cheating

The world of casinos is on the brink of a technological revolution that goes far beyond what we have seen in Hollywood movies about the eye in the sky. At the Global Gaming Expo (G2E) in Las Vegas, it became clear that artificial intelligence (AI) is no longer just a buzzword but is actively intervening in the operations of major resorts. The focus is on monitoring classic table games like blackjack, roulette, and poker. The panel, moderated by Gaming Hall of Fame member Debi Nutton, highlighted that technology is already capable of detecting errors that the human eye simply misses due to the sheer volume of tables.

The operators' goal is clearly defined: increase efficiency, minimize losses, and ensure the guest experience remains undisturbed. Josh Bodell, director of table games for MGM Resorts International, emphasized that integration must be so seamless that players do not even notice the technology. It is no longer just about catching a cheater after minutes of video analysis, but about making split-second decisions. In an industry where every second counts, the speed of data processing is the decisive competitive advantage.

Numbers and facts

The economic necessity of these systems is backed by alarming figures. Trevor Outman, founder and CEO of EagleSight.ai, estimates that casinos lose between 1% and 2.5% of table game revenue every year. The causes are diverse, ranging from simple dealer payoff errors to professional cheating. Since human surveillance has a scalability problem they simply cannot watch every table at once AI steps in here.

EagleSight states that its systems start with an accuracy of about 40% upon introduction but reach 90% within six months as the AI learns from false positives. Resorts World Las Vegas already uses RFID chips in combination with cameras to link individual bets with card recognition. This allows the system to notify staff immediately if a dealer overpays or if a player deviates suspiciously from basic strategy in Three Card Poker. Even at roulette, the technology sends an alert if the ball statistically falls into one section of the wheel too often.

Background

A key aspect of the discussion was the role of the surveillance department. Previously, it was often seen as a necessary cost center. By deploying AI agents, which Trevor Outman categorizes into junior, senior, and orchestrator levels, this area is transforming into a strategic asset. A junior AI agent gathers basic metrics like rounds per hour, while a senior agent actively detects payoff errors. The orchestrator analyzes the entire fleet and recommends that management move staff from slow tables to busy ones.

"We’re transitioning the role from the old perception of surveillance as a cost center to potentially a profit center, increasing the recovery rate." - Trevor Outman, CEO of EagleSight.ai

Skepticism from some guests remains. While some visitors like Susie De Los Santos welcome the technology, others urge caution. MGM CEO Bill Hornbuckle defended the use, comparing data analysis to giants like Amazon. He stated it is about understanding the business better and providing value to customers rather than just viewing them as a data source. The challenge remains the balance between security and privacy.

Why it matters for German players

In Germany, the situation is strictly regulated by the State Treaty on Gambling 2021 (GlüStV 2021). While the AI discussion in Las Vegas primarily concerns land-based casinos, the technology also impacts the online sector where German players are active. Here, AI is already used to ensure player protection. The Joint Gambling Authority of the States (GGL) monitors via systems like LUGAS whether players exceed the monthly deposit limit of 1,000 euros or play at multiple providers simultaneously.

German players should know that automation here primarily serves security. In state-licensed online slots, technology ensures that the one-euro stake limit per spin is strictly enforced. While surveillance in Las Vegas aims to find dealer errors, in Germany, it serves to prevent gambling addiction and comply with legal frameworks. AI could help identify problematic gambling behavior even earlier in the future before financial damage occurs.

What it means for GGL-licensed casinos

For operators with a GGL license, progress in AI surveillance means an opportunity to automate compliance reports. The documentation requirements for the authority in Halle are high. Systems like those presented at G2E could theoretically help report suspicious transactions or gaming patterns in terms of money laundering prevention more quickly. However, data protection hurdles in Germany are significantly higher than in Nevada. German operators must ensure that every form of AI-supported analysis is fully compliant with GDPR and the specific requirements of GlüStV 2021. Only providers on the official whitelist may legally offer their services and must guarantee that player data is secure and not misused for unfair marketing.