Combining Theory of Mind and Kindness for Self-Supervised Human-AI Alignment

TL;DR

This paper introduces a novel approach combining Theory of Mind and kindness to improve self-supervised human-AI alignment, aiming to enhance safety, social intelligence, and value understanding in AI systems.

Contribution

It proposes a new human-inspired framework that integrates Theory of Mind and kindness to better align AI with human values and intentions.

Findings

Enhanced understanding of human mental states

Improved safety and alignment in AI behaviors

Potential reduction in manipulation and harmful actions

Abstract

As artificial intelligence (AI) becomes deeply integrated into critical infrastructures and everyday life, ensuring its safe deployment is one of humanity's most urgent challenges. Current AI models prioritize task optimization over safety, leading to risks of unintended harm. These risks are difficult to address due to the competing interests of governments, businesses, and advocacy groups, all of which have different priorities in the AI race. Current alignment methods, such as reinforcement learning from human feedback (RLHF), focus on extrinsic behaviors without instilling a genuine understanding of human values. These models are vulnerable to manipulation and lack the social intelligence necessary to infer the mental states and intentions of others, raising concerns about their ability to safely and responsibly make important decisions in complex and novel situations. Furthermore, the divergence between extrinsic and intrinsic motivations in AI introduces the risk of deceptive or harmful behaviors, particularly as systems become more autonomous and intelligent. We propose a novel human-inspired approach which aims to address these various concerns and help align competing objectives.