The Elephant in the Room -- Why AI Safety Demands Diverse Teams

TL;DR

This paper proposes treating AI safety and alignment as a social science problem, emphasizing the importance of diverse teams and social science tools to better understand and address AI alignment challenges.

Contribution

It introduces a novel approach to AI alignment that leverages social science methodologies and advocates for diverse teams to improve problem-solving effectiveness.

Findings

Social science tools can be repurposed for AI alignment.

Diverse teams enhance understanding of alignment challenges.

A three-step framework for social science-informed AI alignment.

Abstract

We consider that existing approaches to AI "safety" and "alignment" may not be using the most effective tools, teams, or approaches. We suggest that an alternative and better approach to the problem may be to treat alignment as a social science problem, since the social sciences enjoy a rich toolkit of models for understanding and aligning motivation and behavior, much of which could be repurposed to problems involving AI models, and enumerate reasons why this is so. We introduce an alternate alignment approach informed by social science tools and characterized by three steps: 1. defining a positive desired social outcome for human/AI collaboration as the goal or "North Star," 2. properly framing knowns and unknowns, and 3. forming diverse teams to investigate, observe, and navigate emerging challenges in alignment.