Two kinds of counterfactual closeness

posted in: reading | 0

People often consider counterfactual events that did not happen, and some counterfactuals seem so close to actual events that they are described as aspects of reality. In five pre-registered experiments (N = 1195), we show there are two kinds of … Continued

Aligning AI With Shared Human Values

posted in: reading | 0

We show how to assess a language model’s knowledge of basic concepts of morality. We introduce the ETHICS dataset, a new benchmark that spans concepts in justice, well-being, duties, virtues, and commonsense morality. Models predict widespread moral judgments about diverse … Continued

Regulating LLMs

posted in: reading | 0

This is fairly specific to the AI Act being considered/written in the EU, but the thoughtful discussion is useful for understanding the challenges of regulating emerging AI technologies. Ryan 2302.02337 PDF Document · 396 KB Ryan Watkins, Ph.D.▲ Professor, George … Continued

What We Are Reading: