Request a demo

Find out today the difference that Callsign’s unique solution can make to your business.

Seeing is believing.

General enquiries, support and press

By submitting this form, you agree to Callsign’s Privacy Policy



Thank you for your request


In the meantime, connect with Callsign for insights on authentication and fraud prevention

Back to Knowledge & Insights

Time to jump on the bandwagon? How AI is helping merchants fight fraud

Customer Experience Digital & Commerce Fraud & Scams Payments Behavioural Biometrics

2023 has seen some significant advancements in the world of artificial intelligence (AI) and machine learning (ML). The rise of Chat GPT and other types of generative AI (large language models and chatbots) has made the wider public aware of the ground-breaking potential that such technologies can bring, from streamlining processes to solving customer queries.

For those working across payments, e-commerce and fraud prevention, many are already aware that the benefits AI and machine learning can bring are substantial. For some time now, forms of these technologies have been helping streamline product recommendations and to route customer queries, as well as improving search algorithms. They have also been helping in the fight against fraud.

In this article we’ll examine the benefits that AI can bring to the payments and e-commerce space, and how these technologies are neither complex nor expensive to implement. First, it might be worth defining what we mean by the terms AI and machine learning, as they are often used interchangeably.

  • What is AI: AI can consist of various intelligence signals to enable problem solving through computer science and datasets.
  • What is machine learning: machine learning is a subset of AI through which machines can learn from data without being reprogrammed but instead adapts to the changes. In practice, every form of machine learning involves AI, but not all AI uses machine learning.

For fraud and customer journey practitioners, being able to pull together data from across the organization to inform decision-making is the key to unlocking digital trust with your customers, helping improve user journeys and reducing fraud – and of course, protecting data privacy.

This may seem like an obvious statement in today’s digital world, companies strive to base their decisions on data. All too often though, this data is siloed in different teams or within different tools, rendering it hard to access or making it practically invisible.

Breaking down business silos and AI preconceptions

The benefits that AI and ML can bring is the ability to draw and learn from multiple data sources across a business. In doing so, every team can more effectively communicate and collaborate between each other and different departments. When combined with orchestration tools, these outcomes can be both visualized and the appropriate next steps actioned.

Fraud teams’ data can be enriched with insights ranging from when the customer last accessed the platform, to their spending habits. It’s this level of insight that is helping merchant businesses and acquirers drive down fraud, without impacting the majority of genuine customers as shown in the following examples.

  • Streaming and subscription companies are using data insights such as last login, device type and behavioral insights to detect account sharing activities, using orchestration to relay the appropriate action for their business; whether that’s dynamic message that upsells a premium account, or simply flagging a warning or blocking the user. All without adding friction for genuine users who want to get into their account as quickly as possible.
  • E-commerce businesses are using insights such as spending habits, behavior and device intelligence to detect account takeover, protect loyalty accounts and detect and prevent bots, allowing them to more effectively prevent fraud and streamline authentication journeys.

It isn’t just other teams’ data that can add value. Through these technologies, fraud teams are now easily and cost-effectively able to equip themselves with insights such as keystroke analysis and device fingerprinting – both of which are helping businesses transition away from outmoded solutions such as cookies.

The growing potential of AI and digital technologies in e-commerce

What makes the use of AI and machine learning technologies so valuable is their ability to work at pace. Combining data from across the organization allows for enhanced decision making, helping to detect fraud more accurately whilst allowing genuine customers to progress through login and payment flows more seamlessly.

One example is around device data, and we’ve previously discussed the benefits that device fingerprinting can bring above cookie-based device verification – namely the ability to track across browsers. Organizations are now seeing how this device recognition data can be fed into AI and ML algorithms to leverage enhanced capabilities in detecting and preventing fraud such as account takeover, refund / coupon abuse and first-party misuse – all without adding additional steps to the payment or slowing transaction times.

How does Callsign use artificial intelligence and machine learning?

Callsign uses AI and ML models to identify genuine users via a privacy-preserving digital DNA profile, while simultaneously detecting fraudsters using industry-leading fraud data.

Using industry-recognized artificial intelligence, our unique Dual Synergy methodology layers best-in-class behavioral biometrics with device and location intelligence to build a unique digital profile for every user, while simultaneously detecting fraudsters using industry-recognized fraud models.

In practice, we’re positively identifying the genuine users, which removes false positives from downstream fraud detection. This is done passively by collecting and analyzing a range of device, location, and behavioral data points (as well as any other relevant insights from across the business). Think of it like checking into a hotel ¬¬– once you’ve proved who you are, you don’t need to do it every time you buy a drink or sit down for breakfast.

This means there’s a smaller pool of transactions to analyze, improving operational efficiencies as teams are able to concentrate on fraud fighting efforts, including making it easier to spot fraud.

This approach is helping to drive down the use of SMS OTPs (which for many are a costly hindrance) and help remove the password from payment authentication flows. For those concerned about scalability and new customers, we’ve maintained above 99% acceptance rate during the peak Black Friday period with ≈20% of those being new users.

We know transitioning to new technologies and solutions can be complex for any type of business, so we’re on hand to help offer any advice, tips and guidance in how to approach this transition, or if you’d like a demo.

Get in touch

More Insights

Elevate your scams defense with telco scams models
Orchestration: reacting in real-time
A change in approach for fraud systems?