AI Personalization Evaluation Expert – Multilingual

About this project

Overview

This project focuses on evaluating how well AI assistants can provide helpful, personalized conversation experiences using information available through connected digital services and prior activity.

In this evaluation, you’ll interact with two versions of an AI assistant and compare your experience with each.

What You’ll Be Doing

You’ll start by writing a prompt, then have a natural back-and-forth conversation with two AI assistants before comparing your experiences.

Some questions may connect to information or activity available through your connected services, while others can simply be whatever is on your mind.

What We’ll Assess

After each conversation, you’ll evaluate your experience across several dimensions, including whether the assistant:

• Makes effective use of available information

• Feels like it understands you and your context

• Connects relevant information across the conversation

• Provides responses that feel trustworthy and safe