SafetyDetectives spoke with the co-founder and CEO of Fingerprint, Dan Pinto, about how he got started in cybersecurity, Smart Signals, common techniques used by fraudsters, the role of AI and ML in mitigating fraud attempts, and more.
Can you talk about your background and your current role at Fingerprint?
My tech career began in 8th grade when I scripted bots to automate online video games and sold the resulting in-game items on eBay. After college, I collaborated with a friend to create a search engine for used machinery and equipment, for which I wrote the first version of the crawling and search technology. The first engineer I hired for that company was Valentin Vasilyev, who created the open-source FingerprintJS browser fingerprinting library that became extremely popular. He eventually created the SaaS product known as Fingerprint from his experiences creating that OSS library, and we are once again working together.
I understand you just launched a new product, Fingerprint Pro Plus with Smart Signals. Can you introduce us to the product and explain what Smart Signals is capable of?
Fingerprint Pro Plus with Smart Signals allows customers to pinpoint fraudsters more accurately, even if the visitor is anonymous. Building upon our already market-leading 99.5% accuracy in device identification, Fingerprint’s Smart Signals provide more nuanced information easily integrated into existing fraud models and decision engines via API. These include VPN detection, IP blocklist matching, raw device attributes, Android tamper detection, Android emulator detection and more. We’re also currently working on many more signals to include in future versions. Each signal enhances the ability to target specific use cases, such as account takeover or payment fraud. With the fraud identification confidence provided by Smart Signals, businesses can deliver a frictionless, positive experience for trusted visitors.
What are the key benefits of utilizing Fingerprint’s services for fraud prevention, and how does your company differentiate itself from other fraud detection providers in the market?
Fingerprint Smart Signals discern key traits about a visitor, like a mismatched timezone across different detection techniques, making suspicious users easier to identify. Companies can protect themselves from fraud while still providing trusted users with a personalized experience. Smart Signals provide more real-time data than traditional models, helping Fingerprint customers fortify fraud detection engines and make better, well-informed decisions about their traffic. We put all the data into the hands of our customers.
How does fraud impact businesses, and what are some common techniques and strategies fraudsters use in today’s technology-driven world?
Fraud costs businesses millions of dollars annually, and fraudsters continuously develop more sophisticated techniques. Some of the most common are:
- Omnichannel fraud: Criminals exploit weaknesses in interconnected payment systems and customer accounts.
- Triangulation fraud: Fraudsters act as a middleman between a legitimate seller and a buyer.
- Advertising fraud: A botnet or automated activity executes fraudulent clicks, referrals, malicious redirects or malware installations.
- CNP (card not present) fraud: Hackers steal credit card information to make unauthorized transactions.
- Phishing: Fraudsters employ fraudulent emails and social engineering to steal identities or gain access to accounts.
- Chargebacks: A customer initiates a chargeback with their credit card issuer for a legitimate transaction with the malicious intention of getting a refund while retaining the purchased goods or services.
- Data scraping: Fraudsters automate the extraction of data from websites or online sources to use for nefarious purposes.
Can you provide insights into the latest trends and advancements in analytics that are transforming how organizations derive value from their data?
Analytics advancements allow businesses to gather and understand visitor data quickly. Device intelligence provides additional context to a user’s behavior so companies can promptly flag visitors with malicious intent, preventing fraud before it happens. Analyzing visitor actions can improve user experience by reducing log-in friction, generating targeted advertising and improving account security and privacy.
What role does artificial intelligence and machine learning play in identifying and mitigating security risks and fraud attempts?
Businesses are witnessing exponential data growth — including new browser fingerprinting signals from Fingerprint — and fraud strategies are growing more complex. AI and machine learning help businesses keep up with the continuous data flow and evolving threats by learning normal behavior and identifying anomalies and patterns associated with fraud that human analysis cannot identify. We use machine learning in our own work, and it is frankly amazing what the system can detect and tweak. It’s able to create rules for data subsets that appear inconsequential but, when combined with hundreds of others, expands the frontier of accuracy and effectiveness of our services beyond human abilities. With real-time analysis, companies can identify and prevent potential fraud while reducing false positives for legitimate users. AI can handle large volumes of data, scaling with business growth and providing protection for large enterprises.