[论文解读] General Purpose Artificial Intelligence Systems (GPAIS): Properties, Definition, Taxonomy, Societal Implications and Responsible Governance

Isaac Triguero, Daniel Molina+3 more|arXiv (Cornell University)|Jul 26, 2023

Ethics and Social Impacts of AISocial Sciences参考文献 119被引用 3

一句话总结

本文提出了通用人工智能系统（GPAIS）的全面定义与分类体系，根据自主性、适应性和自我意识，将GPAIS区分为封闭世界与开放世界类型。该文将GPAIS归类为AI驱动的AI与单一基础模型两类方法，指出生成式AI与多模态技术是其关键推动因素，并倡导建立负责任的治理与可信性框架，以确保这些系统的安全、可控部署。

ABSTRACT

Most applications of Artificial Intelligence (AI) are designed for a confined and specific task. However, there are many scenarios that call for a more general AI, capable of solving a wide array of tasks without being specifically designed for them. The term General-Purpose Artificial Intelligence Systems (GPAIS) has been defined to refer to these AI systems. To date, the possibility of an Artificial General Intelligence, powerful enough to perform any intellectual task as if it were human, or even improve it, has remained an aspiration, fiction, and considered a risk for our society. Whilst we might still be far from achieving that, GPAIS is a reality and sitting at the forefront of AI research. This work discusses existing definitions for GPAIS and proposes a new definition that allows for a gradual differentiation among types of GPAIS according to their properties and limitations. We distinguish between closed-world and open-world GPAIS, characterising their degree of autonomy and ability based on several factors such as adaptation to new tasks, competence in domains not intentionally trained for, ability to learn from few data, or proactive acknowledgment of their own limitations. We propose a taxonomy of approaches to realise GPAIS, describing research trends such as the use of AI techniques to improve another AI (AI-powered AI) or (single) foundation models. As a prime example, we delve into GenAI, aligning them with the concepts presented in the taxonomy. We explore multi-modality, which involves fusing various types of data sources to expand the capabilities of GPAIS. Through the proposed definition and taxonomy, our aim is to facilitate research collaboration across different areas that are tackling general purpose tasks, as they share many common aspects. Finally, we discuss the state of GPAIS, prospects, societal implications, and the need for regulation and governance.

研究动机与目标

• 解决当前缺乏清晰、可扩展的通用人工智能系统（GPAIS）定义的问题，以涵盖不同自主性与能力水平的系统。
• 基于底层机制，提出GPAIS的分类体系，区分AI驱动的AI与单一基础模型。
• 分析生成式AI与多模态技术在提升GPAIS能力方面的作用。
• 识别GPAIS的开放性挑战与社会影响，尤其关注安全性、控制力与可信度问题。
• 倡导建立负责任的治理与监管框架，以确保GPAIS的可信与可控发展。

提出的方法

• 提出一种新的GPAIS定义，强调少样本学习、迁移学习以及对自身局限性的主动自我意识等特性。
• 引入一种分类体系，区分封闭世界GPAIS（适应能力有限）与开放世界GPAIS（高度自主与自我改进能力）。
• 将GPAIS划分为两大主要类别：AI驱动的AI（利用AI设计或优化其他AI系统）与单一基础模型（在广泛数据上训练的单体模型）。
• 分析生成式AI与多模态系统作为GPAIS的关键实现形式，强调其在实现零样本与少样本泛化中的作用。
• 将GPAIS的可信性映射至NIST人工智能风险管理框架，识别出七个维度（如安全性、公平性、可解释性）下的150项可信属性。
• 提出治理机制，包括对开发者的法律问责制，以及基于模型能力的安全标准。

实验结果

研究问题

• RQ1 如何形式化定义GPAIS，以准确捕捉其自主性与泛化能力的差异？
• RQ2 构建GPAIS的关键技术路径是什么？它们在架构与能力上如何区别？
• RQ3 当前的基础模型与生成式AI系统在多大程度上展现出超越训练分布的真正泛化能力？
• RQ4 如何使GPAIS具备可信性、安全性与可控性，特别是在开放世界场景中？
• RQ5 需要哪些监管与治理框架，以确保GPAIS的负责任发展与部署？

主要发现

• 所提出的GPAIS定义能够基于适应性、自我意识与少样本学习能力，对封闭世界与开放世界系统做出细致区分。
• 该分类体系识别出GPAIS的两大主要发展路径：AI驱动的AI（多AI系统协同）与单一基础模型（单体、多任务学习模型）。
• 生成式AI与多模态系统被证明是推动GPAIS发展的核心因素，尤其体现在对多样化输入的零样本与少样本泛化能力上。
• 当前的GPAIS虽未达到AGI水平，但已展现出推理、适应与任务迁移等涌现能力，尽管仍缺乏完全类人意识或主动性。
• 治理框架至关重要，需通过开发者的法律问责制与基于模型能力的安全标准，防范可预见的危害。
• 可信性可通过将150项属性系统性地映射至NIST AI风险管理框架来实现，涵盖安全性、公平性、透明性与韧性等方面。