Displaying 14 resources

A General Language Assistant as a Laboratory for Alignment
Given the broad capabilities of large language models, it should be possible to work towards a general-purpose, text-based assistant that is aligned with human values, meaning that it is helpful, honest, and harmless.

Ensuring the Safety of Artificial Intelligence
The essays are some of our first steps towards an understanding of how to make today’s choices in ways that take the people of tomorrow seriously. This is not an easy undertaking.