Implicit Values Embedded in How Humans and LLMs Complete Subjective Everyday Tasks
This addresses the problem of value misalignment in AI assistants for users relying on them for subjective everyday tasks, highlighting a critical gap in current systems.
The study investigated the alignment of implicit values, such as environmentalism and charity, between six popular LLMs and 100 human crowdworkers when completing 30 everyday tasks, finding that LLMs often do not align with humans or each other in these values.
Large language models (LLMs) can underpin AI assistants that help users with everyday tasks, such as by making recommendations or performing basic computation. Despite AI assistants' promise, little is known about the implicit values these assistants display while completing subjective everyday tasks. Humans may consider values like environmentalism, charity, and diversity. To what extent do LLMs exhibit these values in completing everyday tasks? How do they compare with humans? We answer these questions by auditing how six popular LLMs complete 30 everyday tasks, comparing LLMs to each other and to 100 human crowdworkers from the US. We find LLMs often do not align with humans, nor with other LLMs, in the implicit values exhibited.