Skip to contentSkip to navigationSkip to topbar
Twilio Docs
On this page

Fraud Guard AI Nutrition Fact Labels

The AI Nutrition Facts Label is intended to give consumers and businesses a more transparent and clear view into "what's in the box" of a specific AI model. It explains how their customer data is being used, especially when it comes to training AI models.

Twilio Verify Fraud Guard is built on Twilio's own proprietary AI model. Below are the current AI Nutrition Fact Labels for Twilio's AI Base Model.

For Twilio's own AI model, customer meta data is used to train its Base Model.

Please visit Twilio AI Nutrition Facts(link takes you to an external page) to learn more about AI Nutrition Facts and how to read these labels.

Twilio's own AI model

twilios-own-ai-model page anchor

AI Nutrition Facts

Verify Fraud Guard

Description
Fraud Guard detects and prevents SMS pumping abuse in real time to protect customers from artificially-inflated traffic using first of its kind technology and Twilio's Proprietary AI Model engine.
Privacy Ladder Level
2
Feature is Optional
Yes
Model Type
Predictive
Base Model
Twilio Proprietary Model + Prophet (only used in Fraud Guard's Standard & Max Protection Modes)

Trust Ingredients

Base Model Trained with Customer Data
Yes

Customer messaging traffic meta data is used for model training.

Customer Data is Shared with Model Vendor
No
Training Data Anonymized
Yes
Data Deletion
Yes
Human in the Loop
Yes
Data Retention
30 days

Compliance

Logging & Auditing
Yes

Standard service logging is applied and logs are stored for future review

Guardrails
N/A
Input/Output Consistency
Yes
Other Resources
Verify Fraud Insights, Verify Geo Permissions, Global Safe List APIs,

Learn more about this label at nutrition-facts.ai(link takes you to an external page)