The complexities of state-sponsored terrorism, professional criminals, and basement bad actors are becoming more difficult to comprehend, track, expose, and prevent. In today’s world, detecting fraud requires a comprehensive approach that matches data points with activities to determine what is abnormal.
An international tech giant behemoth has unveiled details of its upcoming AI-powered chip designed to bring deep learning inference to enterprise workloads to help address fraud in real-time, to identify and stop a variety of fraud attacks and crime quickly and accurately – while improving customer and citizen experiences.
The AI chip includes on-chip acceleration for AI inference during a transaction. The breakthrough of the new on-chip hardware acceleration, which took three years to develop, is intended to help customers achieve business insights at scale across banking, finance, trading, insurance applications, and customer interactions.
According to recent research commissioned by the tech company, 90 % of respondents said it is important to be able to build and run AI projects regardless of where their data resides. According to the company, the technology is intended to allow applications to run efficiently regardless of where the data resides, assisting in overcoming traditional enterprise AI approaches that tend to require significant memory and data movement capabilities to handle inferencing.
Businesses in the Philippines require this technology support; in 2017, an internal fraud involving the theft of 1.75 billion pesos ($34.5 million) at a major Philippine lender was revealed, in the latest controversy to afflict the Philippine banking sector. According to reports, the bank discovered the fraud after a client denied taking out the loans. During the coronavirus disease 2019 (Covid-19) pandemic, there was an increase in digital fraud attempts against businesses and consumers in the Philippines.
“With this tech, the accelerator close to mission-critical data and applications means that enterprises can conduct high volume inferencing for real-time sensitive transactions without invoking off-platform AI solutions, which may impact performance. Clients can also build and train AI models off-platform, deploy and infer on a system for analysis,” the company said.
As per the company, businesses typically use detection techniques to detect fraud after it has occurred, a process that can be time-consuming and compute-intensive due to the limitations of today’s technology, especially when fraud analysis and detection is performed far away from mission-critical transactions and data.
“Due to latency requirements, complex fraud detection often cannot be completed in real-time – meaning a bad actor could have already successfully purchased goods with a stolen credit card before the retailer is aware fraud has taken place,” it said.
The new chip has an innovative centralised design that enables clients to use the full power of the AI processor for AI-specific workloads, making it ideal for financial services workloads such as fraud detection, loan processing, trade clearing and settlement, anti-money laundering, and risk analysis. The chip has 8 processor cores with a deep superscalar out-of-order instruction pipeline and a clock frequency of more than 5GHz, which is optimised for the demands of heterogeneous enterprise-class workloads.
“With these innovations, clients will be positioned to enhance existing rules-based fraud detection or use machine learning, accelerate credit approval processes, improve customer service and profitability, identify which trades or transactions may fail, and propose solutions to create a more efficient settlement process,” said the tech company.
Fraud detection and prevention is not a one-time event. There is no beginning or endpoint. Rather, it is a continuous cycle of monitoring, detection, decisions, case management, and learning that feeds detection improvements back into the system. Organisations should strive to learn from fraud incidents continuously and incorporate the findings into future monitoring and detection processes. As artificial intelligence and machine learning become more common, the next generation of technologies is automating manual processes associated with combining large data sets and utilising behavioural analytics.
