“We cannot outright predict a pandemic any more than we can predict a data breach or an account takeover,” DataVisor CEO and Co-Founder Yinglian Xie recently told PYMNTS. “What we can do, however, is empower organizations to proactively spot burgeoning crises early and take decisive action before extensive damage occurs.” Learn how DataVisor uses artificial intelligence (AI) and unsupervised machine learning to mitigate everyday fraud, as well as very rare occurrences, in Black Swan from PYMNTS.
The following is an excerpt from Black Swan, contributed by DataVisor CEO and Co-Founder Yinglian Xie.
AI And Machine Learning Can Help Us Proactively Prepare For The Unknown
As a global community, we have made tremendous strides in improving our ability to respond to major crises. Despite our progress, however, we still struggle in the face of the unexpected.
Today, we are dealing with the stark realities of COVID-19 and experiencing anew how failure to adequately prepare for the unknown can result in tragedy. The truth is, we cannot predict the onset of a global pandemic. However, history teaches us that, with hindsight, we can often pinpoint the telltale signs. Therein lies our hope for the future. We can deconstruct the anatomy of past crises and threats and apply that intelligence to the defenses we build today as we prepare for the challenges of tomorrow. Viral attacks in the human world operate similarly to viral attacks in the digital realm, and what we learn from one can help us understand the other.
We are fortunate to have the technological means to perform so many essential functions online. However, the more we transact online, the more vulnerable we are to digital threats. This is especially the case when it comes to digital transactions and payments.
Banks are closing branches and encouraging app use instead. Big-box stores are limiting hours, recommending payment apps at checkout and promoting online shopping. People are relying on services like Apple Pay, PayPal, Stripe and Venmo for payments. Simultaneously, banks are encouraging people to open new online accounts, take advantage of special promotions and leverage online services. In short, there is a dizzying blur of digital activity happening across the global economy — and where there is chaos at scale, you can be sure fraudsters are close by.
Fortunately, we have the technological capabilities to analyze, process and derive actionable insights from all this data, and we can do so in real time using AI and unsupervised machine learning. Even as fraudsters leverage the speed of automation and the scale of vast bot armies, we have solutions that can surface correlated patterns and expose hidden connections. From mass registrations of fake new accounts to promotion abuse attacks to account takeovers, we can identify where suspicious activity is taking place by holistically analyzing raw data, without the need for labels or pre-existing rules.
We cannot outright predict a pandemic any more than we can predict a data breach or an account takeover. What we can do, however, is empower organizations to proactively spot burgeoning crises early and take decisive action before extensive damage occurs.
A recent article from WIRED highlights how lessons learned from SARS and H1N1 helped Hong Kong, Taiwan, Japan and South Korea respond swiftly, successfully and proactively to COVID-19. When it comes to black swan events in the digital economy, we can do the same. By combining what we already know with new AI-powered capabilities to process the unknown in real time, we can architect defense and response systems to mitigate damage associated with major threat attacks.
Read more insights and tips in Black Swan from PYMNTS.