# Aligned with Whom? Direct and social goals for AI systems

arXiv: https://arxiv.org/abs/2205.04279 · HTML: https://mlsanity.com/paper/2205.04279/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 28 references that are in the map.
- First posted on arXiv 9 May 2022.

## Builds on (most cited first, 4 of 4)

- [Artificial Intelligence, Values and Alignment](https://mlsanity.com/paper/2001.09768/) (2020)
- [Open Problems in Cooperative AI](https://mlsanity.com/paper/2012.08630/) (2020)
- [Impossibility and Uncertainty Theorems in AI Value Alignment (or why your AGI should not have a utility function)](https://mlsanity.com/paper/1901.00064/) (2018)
- [Agents and Devices: A Relative Definition of Agency](https://mlsanity.com/paper/1805.12387/) (2018)

## Built on by (most cited first, 7 of 7)

- [A Survey on Personalized Alignment -- The Missing Piece for Large Language Models in Real-World Applications](https://mlsanity.com/paper/2503.17003/) (2025)
- [Mapping the Ethics of Generative AI: A Comprehensive Scoping Review](https://mlsanity.com/paper/2402.08323/) (2024)
- [Academically intelligent LLMs are not necessarily socially intelligent](https://mlsanity.com/paper/2403.06591/) (2024)
- [From Anger to Joy: How Nationality Personas Shape Emotion Attribution in Large Language Models](https://mlsanity.com/paper/2506.02431/) (2025)
- [Traccia: An OpenTelemetry-Based Governance Platform for AI Systems](https://mlsanity.com/paper/2607.14309/) (2026)
- [Dialogical Reasoning Across AI Architectures: A Multi-Model Framework for Testing AI Alignment Strategies](https://mlsanity.com/paper/2601.20604/) (2026)
- [Demographic Prompting at Scale: When More Attributes Hurt LLM--Human Agreement](https://mlsanity.com/paper/2607.10590/) (2026)

## Abstract (arXiv)

As artificial intelligence (AI) becomes more powerful and widespread, the AI alignment problem - how to ensure that AI systems pursue the goals that we want them to pursue - has garnered growing attention. This article distinguishes two types of alignment problems depending on whose goals we consider, and analyzes the different solutions necessitated by each. The direct alignment problem considers whether an AI system accomplishes the goals of the entity operating it. In contrast, the social alignment problem considers the effects of an AI system on larger groups or on society more broadly. In particular, it also considers whether the system imposes externalities on others. Whereas solutions to the direct alignment problem center around more robust implementation, social alignment problems typically arise because of conflicts between individual and group-level goals, elevating the importance of AI governance to mediate such conflicts. Addressing the social alignment problem requires both enforcing existing norms on their developers and operators and designing new norms that apply directly to AI systems.