Real-time fraud detection with Feature Views
Real-time fraud detection depends on three kinds of features: what a customer did...
Real-time fraud detection depends on three kinds of features: what a customer did...
BackgroundAccording to the Communications Fraud Control Association's 2025 survey...
Author: Fang Fang, Machine Learning & Decision Platform, AffirmContinue reading on Affirm Tech Blog »
Most anti-fraud architectures in enterprise payment workflows evaluate transactions in isolation. When an invoice arrives, an automated rules engine checks it against static thresh...
How one reflection exposed a 500-account fraud ring and why cross-industry signals matter.
During International Fraud Awareness Week, Eisner's Scott Moritz, a former FBI special agent, lays out the pervasiveness of fraud, points out some important red flags — and offers...
Understanding How Machine Learning-Based Fraud Detection Systems Work, Their Types, Key Technologies, Real-World Applications, Benefits…Continue reading on Medium »
Learn how Snowflake ML powers real-time fraud detection with the Online Feature Store (now GA), low-latency model serving and integrated MLOps — all on a single, secure platform. S...
The U.S. department has increased its scrutiny of payments using new sources of data in an attempt to stem past ‘pay and chase’ patterns.
Every business with a checkout loses revenue it never counts. The fraud report shows how many transa...
Most fraud detection tutorials stop at “train a model and print the accuracy.” That’s the easy 10%. The other 90% — the part that decides…Continue reading on Medium »
Cifas recorded 220,041 fraud-risk cases in the first half of 2026, just 1% up on the same period las...
A payment can move between accounts before a person has time to review it. Real-time AI fraud detection helps financial institutions assess transactions as they arrive, using patte...
What if we stopped thinking about fraud as simply a criminal problem and started thinking about it as a quality problem?
Built for the TigerGraph Agentic Fraud Investigation challenge (Hacker House Goa, 2026)Continue reading on Medium »
Just 7% rate their organizations more than moderately prepared for AI-powered fraud
AI fraud detection now hinges on data quality, not model sophistication, because fragmented data drives false positives that erode client trust.
Author: A finextra community member
The fraud detection technologies used in insurance claims today center on AI-driven pattern scoring, multimodal AI for images and documents, network and social-graph analysis, natu...
Professional fraud rings spread their schemes across insurers. Here’s what carriers should test before trusting a vendor’s fraud network or building their own
In Part 3 of our sentiment series, we ended on a cautious note: a model can be highly confident and still be wrong. This article puts a number on that caution. We generate 500 synt...
The fraud risk the industry is least equipped to handle, and the payments reality most people in insurance have wrong
Just imagine you have booked a cab, and the driver asks you to cancel the ride because the fare shown is very low and asks you to take a private ride with them. Such fraudulent ac...
Digital commerce faces a quiet but expensive operational crisis. Global card-not-present fraud losse...
Use SocialBu to discover ideas, generate post drafts, and schedule them across your social channels.