World Digital Technology Academy

Research & Publication

Latest Research

Advancing knowledge through innovation and collaboration.

Single AI Agent Runtime Security Testing Standards

The "Single AI Agent Runtime Security Testing Standards" aim to establish a framework and methodology for evaluating the safety and security of single-agent systems during runtime. This standard seeks to provide organizations with a unified approach to identifying and mitigating risks, ensuring the safety, reliability, and controllability of intelligent agents in real-world applications.

Nov/22/2025

Generative AI Application Security Testing and Validation Standard

This standard document provides a framework for testing and validating the security of Generative AI applications. The framework covers key areas across the AI application lifecycle, including Base Model Selection, Embedding and Vector Database in the Retrieve Augment Generation design patterns, Prompt Execution/Inference, Agentic Behaviors, Fine-Tuning, Response Handling, and AI Application Runtime Security.

Dec/09/2025

Cart

Single AI Agent Runtime Security Testing Standards

Generative AI Application Security Testing and Validation Standard