English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We explore the idea of aligning an AI assistant by inverting a model ofusers' (unknown) preferences from observed interactions. To validate ourproposal, we run proof-of-concept simulations in the economic ultimatum game,formalizing user preferences as policies that guide the actions of simulatedplayers. We find that the AI assistant accurately aligns its behavior to matchstandard policies from the economic literature (e.g., selfish, altruistic).However, the assistant's learned policies lack robustness and exhibit limitedgeneralization in an out-of-distribution setting when confronted with acurrency (e.g., grams of medicine) that was not included in the assistant'straining distribution. Additionally, we find that when there is inconsistencyin the relationship between language use and an unknown policy (e.g., analtruistic policy combined with rude language), the assistant's learning of thepolicy is slowed. Overall, our preliminary results suggest that developingsimulation frameworks in which AI assistants need to infer preferences fromdiverse users can provide a valuable approach for studying practical alignmentquestions.

Social Contract AI: Aligning AI Assistants with Implicit Group Norms