Safe and responsible development with Generative Language Models
Guidance for responsible development of applications with large language models (LLMs), from planning to testing.
Safe and responsible development with Generative Language Models
Guidance for responsible development of applications with large language models (LLMs), from planning to testing.
/2023/data/im/04c03aa2-e9ac-44f8-865a-33c1a494e11d.jpeg
Overview
Generative language models are an incredible new technology that enable computers to converse about almost any topic. Unfortunately, we don't always know what those models are going to say. How do we ensure that our applications don't produce inappropriate content? In this talk, we'll explore guidance for curbing those risks and developing safe and responsible LLM-backed applications.
Speakers
Resources
Google AI Principles
Learn moreRAI Toolkit in TensorFlow
Learn morePaLM API and MakerSuite Safety Controls
Learn more
I/O Connect
Join sessions, office hours, and demos at Google's in-person developer events.
I/O Extended
Celebrate I/O technology with a local community-led event near you.