The following article is provided by Mark Goldspink, CEO of The ai Corporation, providers of fraud and risk management services.
Artificial intelligence (AI) will revolutionise the ways businesses work
The growing focus on customer experience means businesses now have no choice but to continuously improve their customer journey. AI is central to brands taking digitalisation to a new, exciting level. Ensuring they are accessible across all channels has become key and artificial intelligence is helping organisations boost accessibility and their bottom line.
AI is already impacting the business landscape in many ways: biometrics and facial recognition technology has meant companies can perform real-time comparisons to image databases. It is being used to increase personalisation and drive deeper relationships between brands and their customers and we are seeing the use of natural language and voice-based user interfaces across e-commerce.
Even the data collected by fraud platforms is being used for more than just identifying fraud. The data from fraud platforms can be utilised in many ways, for ‘good’, as well as ‘bad’, for example, in analyzing spending patterns amongst customer data and helping marketing teams to develop targeted marketing campaigns. Purchasing data can also help a brand to identify customer segments and establish target markets for advertising.
So, what is AI and how can it help?
In simple terms, artificial intelligence is a platform that appears to be intelligent, and which can exceed the performance of humans. It is a broad description of any device that mimics human or cognitive functions such as reasoning or problem solving.
Machine learning (ML) is one example of AI. It is a statistical and data driven approach to creating AI, for example, when a computer program learns from data to improve its performance in completing a task. ML starts off by making lots of mistakes, the machine then learns from these mistakes and improves its performance on future tasks. Learning from historical data in this way is the most successful approach to generating many different types of AI, from visual recognition to spam filtering.
“I like to think of ML as a cog of AI,” The ai Corporation CTO Tom Myles said. “When a computer with ML is tasked with learning a problem, the first thing it will do is observe historical data so it can adapt its own processes and understanding, to predict future occurrences of a scenario. A computer that has learned something from data in this way can be considered intelligent and so is an example of AI.”
Progression of technology means that traditional AI devices are no longer considered AI. In the 1960s, Hollywood imagined AI to be talking robots or machines that serve our needs. While we now have commodities such as Siri and Alexa which are deemed clever, they are no longer AI. This is because of the AI effect. Some people believe AI is ‘whatever hasn’t been done yet’. Another example would be large rule sets being processed in milliseconds that mimic human assessments (and so is strictly AI) but such tasks are now standard computerized tasks (so no longer appear intelligent because we can comprehend).
When AI is successful at a task, its processes become comprehensible. This means the device is not really ‘intelligent’ (in the human sense) and the device is no longer considered AI – essentially, after revealing the magician’s trick, it is no longer magic.
Man & machine – Mitigating risk, however, remains imperative to success
Research into behavioural economics shows that as humans, certain biases (decision fatigue, intentional blindness, and herd instinct) impact our ability to operate perfectly. Expecting fraud teams to work through thousands of pieces of data and be spot on every time, is simply not practical. Machines need to be allowed to lead some stages of the tedious, repetitive processes – releasing human creativity.
Using best of breed machine learning technology, our customers have significantly reduced the amount of time it takes to analyse data and have provided increased accuracy in fraud detection and a reduction in false positive rates, meaning less declines and more transactions.
The role of the human, interpreting, analysing and understanding is still key to the business process. I believe that there is a role for both humans and machines in the world of business. Instead of man versus machine, I believe the approach should be man AND machine.
There are many factors driving the requirements need for effective AI solutions for payments and transaction processing. Firstly, as technology evolves, online fraud is becoming more prevalent and damaging, with financial services and e-commerce companies especially vulnerable to attacks.
Modern fraudsters have evolved their ability to detect vulnerabilities in systems, and are shifting their targets to those weak links. They employ new tactics too – using distributed networks, big data and the dark web to locate vulnerabilities and maximise the associated risk. Fraudsters are also devising multidimensional tactics that inflict damage by sequentially compromising more than one point of vulnerability.
AI, more specifically ML, is already helping organizations combat fraud in ways that just weren’t possible previously. It is an exciting time for businesses, with disruptive opportunities in virtually every market sector.
Organisations that want to defend themselves against these risks and thwart modern fraud attacks must be able to react in real time. To do this, they need powerful solutions that are responsive and dynamic, yet still easy to use and integrate into their existing systems.
Traditional rules-based fraud management engines are breaking down at this level of sophistication, speed and scale. What is needed is a paradigm shift in the tools used to fight multichannel commerce and banking fraud. AI solutions can replace high-maintenance, rules-based fraud management tools with self-learning algorithms, reducing ‘false positives’ by using big data to identify new fraud patterns. Ultimately, these capabilities enable managers to make better decisions related to fraud, and so significantly reduce fraud loss.
After graduating from London University with a PhD In Chemistry, Mark started worked for a multinational oil company Texaco for 12 years and was mainly involved in forecourt retailing.
In 2000 Mark moved to an internet billing company and in 2005 became Managing Director for Retail Decision’s global payments and fraud Division. In 2010, he moved to Logica (bought by CGI Inc. 2012) and was vice president responsible for managing Shell’s outsourced payment contracts worth over $500 million. He joined ai in 2013 to work with Ashley Head on developing and expanding a whole series of inter-related payment businesses globally.
Founded in 1998, The ai Corporation (ai) has a long and exciting heritage as one of the world’s leading companies in fraud, risk management and pioneering business intelligence from payments data. In 2016 ai purchased a payment gateway and now also offers all its solutions via managed services. For more than 20 years, ai has provided solutions to some of the world’s largest financial institutions, international merchants and other major payment service providers. Today, our fraud detection solutions, including our new ‘state of the art’ neural technology, protect and enrich payments experiences for more than 100 banks, more than three million multi-channel merchants and more than 300 million consumer cardholders. Find out how we can help your organisation at www.aicorporation.com .
Like this article? Take a second to support us on Patreon!