Whose Truth? Pluralistic Geo-Alignment for (Agentic) AI

TL;DR

This paper highlights the importance of geographic and cultural context in AI alignment, emphasizing the need for spatio-temporally aware approaches to ensure AI systems behave appropriately worldwide.

Contribution

It introduces the concept of pluralistic geo-alignment, reviews key research challenges, and proposes methods for assessing geographic sensitivity in AI alignment.

Findings

AI outputs vary significantly across regions due to cultural norms.

Current alignment methods may not account for geographic differences.

Automated, transparent geo-alignment is crucial for global AI deployment.

Abstract

AI (super) alignment describes the challenge of ensuring (future) AI systems behave in accordance with societal norms and goals. While a quickly evolving literature is addressing biases and inequalities, the geographic variability of alignment remains underexplored. Simply put, what is considered appropriate, truthful, or legal can differ widely across regions due to cultural norms, political realities, and legislation. Alignment measures applied to AI/ML workflows can sometimes produce outcomes that diverge from statistical realities, such as text-to-image models depicting balanced gender ratios in company leadership despite existing imbalances. Crucially, some model outputs are globally acceptable, while others, e.g., questions about Kashmir, depend on knowing the user's location and their context. This geographic sensitivity is not new. For instance, Google Maps renders Kashmir's borders differently based on user location. What is new is the unprecedented scale and automation with which AI now mediates knowledge, expresses opinions, and represents geographic reality to millions of users worldwide, often with little transparency about how context is managed. As we approach Agentic AI, the need for spatio-temporally aware alignment, rather than one-size-fits-all approaches, is increasingly urgent. This paper reviews key geographic research problems, suggests topics for future work, and outlines methods for assessing alignment sensitivity.