How casinos can utilize AI
Source: John Liu
The use of artificial intelligence (AI)
to improve casino security is expected to increase in coming years. But for now, video analytics
is where AI is likely to be applied most extensively.
“Artificial intelligence is driving the next generation of video analytics,” said AJ Frazer, VP of Business Development at Agent Video Intelligence
, whose solutions are employed by several casinos.
Deep learning, wherein a machine processes large amounts of data to make its own intelligent judgements, is an essential part of AI for video analytics. In general applications, video analytics today are trained to identify predefined anomaly events that require security staff’s attention.
David Green, Principal of the Macau-based Newpage Consulting
, pointed out that analytics have always been important for casinos. “Game results are analyzed to look for statistically abnormal variances which may suggest cheating, collusion or training deficiencies. They may also point to mechanical or electrical equipment failures which players can exploit.”
Through deep learning, surveillance systems are able to spot suspicious activities like thefts without the need for casino operators to manually predefine anomalies. “The analytics system learns the normal behavior patterns for different times of the day and different days of the week, and then alerts on any behavior which deviates from the norm,” Frazer explained.
Agent Video Intelligence said it had installed casino solutions to alert operators to unusual events and/or behavior patterns “that would otherwise go unnoticed.”
“Millions of images extracted from real-life surveillance video footage” contribute to “unparalleled detection accuracy and extremely low false alarm rates,” Frazer said.
Big data extracted from video could also be aggregated with other sources to provide casinos with better, more accurate insights about behavior trends that over time could enable proactive security measures, Frazer said.
Kok Tin Gan, a partner in PwC’s Cybersecurity & Privacy
practice, points out that with InidgoVision’s solutions, casinos can track different colors, directions, sizes and speeds of individuals and analyze their routes and dwell times.
Casinos can also now easily identify blacklisted individuals within a crowd thanks to cameras with deep learning-based facial recognition capability.
“With the help of an AI-based detective system, security staff can easily detect metallic and nonmetallic explosives, firearms and other weapons concealed on an individual,” Gan said.
AI helps to secure casino property
AI and big data have broader applications in relation to casino security. They can be applied in employee screening, security staff training, tracking of chips and cash, and data and infrastructure security, according to Gan.
The idea of AI or deep learning is for machines to make intelligent judgments on their own. AI may be deployed to monitor network infrastructure as well as data security “to proactively detect any malicious actions such as transferring a large amount of personal information out from a casino data network infrastructure,” Gan said.
There are also several security vendors (e.g. Cisco, Darktrace, Forcepoint) that leverage deep learning to detect potentially malicious activities in a casino’s IT infrastructure.
AI for human resources
“By using natural language processing and machine learning, casinos can leverage a global set of data from social media (Linkedin, Facebook, Twitter) and job searching websites to perform a background check with higher accuracy in a short period of time,” Gan said.
Once a new employee is hired, casinos can leverage AI to facilitate the employee orientation process through a tailor-made new-hire training program or chat bot to provide answers to some of the most commonly asked questions.
Despite the promise of AI for casinos, deployment is still at a nascent stage.
Alfred Lee, Senior Director at Surveon Technology, which provides various types of surveillance cameras for casinos including the Phoenix Casino in Vietnam, pointed out that “many companies, including us, are still studying how to apply AI in casinos.”
While the future looks bright for AI in casinos, the technology is still fairly new and with this comes challenges. As a result, many of potential applications of in casinos are still in the planning stage.