When the Cloud Decides: Designing for predictive machine learning for the IoT

Increasingly, predictive machine learning drives the consumer Internet of Things. Cloud-based algorithms collect data from thousands of connected devices, then create models that aim to predict what behavior will create the most positive outcome. But how should the human users, the supposed beneficiaries of these products and services, interact with these algorithms? How do I change the behavior of a predictive lawn care system when my only interface to it is a garden hose? This presentation will outline the challenges of creating embedded intelligent systems, and will offer some UX design approaches for addressing these challenges.