Overview
In this session, we will guide you through evaluating the safety and biases of GenAI models. Whether you're a model builder, app developer, or industry regulator, this session will equip you with a clear, risk-based approach to identifying and addressing key safety concerns.. Here's what you can expect:
A bit of theory: Why AI safety matters, an overview of safety methods, and key safety evaluation techniques.
DIY guide: How to create your own safety policy, evaluate fairness and biases, and when to apply red-teaming.
What to assess: Major risk categories, user and model intents, and critical scenarios to watch for.
Speakers
Ilya Kochik
Toloka
VP of Business Development
Alexander Borodetskiy
Toloka
VP of Growth,
Responsible AI