Breaking

Nace AI Launches Drex for Finance, a Model Built for Rapid Financial Decisions

November 2, 2026 at 13:35 EST

Finance

Nace AI has launched Drex for Finance, a model designed for split-second financial tasks such as judging whether a card payment looks fraudulent.

Product launch · Nace AI

Drex for Finance: a model that scores fixed answers in split-second fraud decisions

Nace AI's new model is built to choose among predefined outcomes, such as approve, flag, or decline, and to return a probability for each option. The design aims at real-time payment checks, where speed and graded signals matter.

How a graded fraud decision works

Flow as described in the launch, not measured performance

STEP 1

Inputs

Transaction log plus supporting documents

→

STEP 2

Fixed answer set

Model selects among predefined outcomes instead of writing free text

→

STEP 3

Probability per option

A graded score for each candidate answer

→

STEP 4

Institution sets thresholds

Block, review, or allow each transaction

Why it may matter

Built for real-time decisions in a short approval window
Probability scores let each institution choose its own cutoffs
Predictable, fixed output formats suit regulated, auditable workflows

What remains unproven

No independent benchmark results were published
No accuracy or false-positive rates were provided
No comparison yet against rule-based or ML fraud systems

Buyer checklist

A probability score is only as useful as its calibration and its performance on real, imbalanced transaction data.

Calibration

Do scores reflect true fraud likelihood?

False positives

How many legitimate payments get blocked or reviewed

Real-world data

Performance on rare-fraud, imbalanced transaction sets

Cost and adoption

Pricing and which institutions use it are not yet disclosed

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.