QA for AI: Testing Machine Learning Applications - The Reality of Developing an Artificial World

Explore the challenges and strategies for quality assurance in AI-driven applications in this 54-minute conference talk. Delve into the critical importance of testing machine learning-enabled systems, gaining insights into testable ML features and a step-by-step approach to ensure AI applications function as intended. Learn from real-world examples, including the Apple Credit Card controversy, and understand the ethical implications of AI development. Discover the "Pox Method" for testing machine learning algorithms, and examine key considerations such as diversity, security, privacy, and bias in AI systems. Gain valuable knowledge on testing for outliers and ensuring the quality of cutting-edge AI applications across various industries.

Introduction
Scenario
Apple Credit Card Story
Diversity in Tech
Testing Recommendations
IBM Watson
Testing Machine Learning Algorithms
Learning How Machine Learning Works
Testing Machine Learning
The Pox Method
The Problem
The Bug Report
The Meeting
The Trolley Problem
Ethics
Diversity
Security Privacy Bias
QA for AI
Testing for outliers