7.6 Artificial Intelligence (AI)

Program‎ > ‎

7.6 Artificial Intelligence (AI)

7.6A AI - Introduction

Class Outline

Intelligence and Emotion

Deep Learning, Machine Learning, and AI

Autonomous Systems

Superintelligence

(Applications)

Class Description (by Jaeseung Jeong)

[TBC]

Lecture Candidates: Jaeseung Jeong, Joanna Bryson/Bath

7.6B AI Governance

Class Outline (Draft)

AI Principles; Asilomar Principles, Asimov's Law,

Ethics

Safety

Regulation

Impact on Labor Market

Class Description (by Danit Gal)

This lecture on AI governance will be divided into three sections. The first will provide an overview of the ethical, inclusive, and societal aspects of AI. We will examine key documents such as the IEEE’s Ethically Aligned Design v.2, the Asilomar Principles, and the AI Now report. The lecturer will also offer first-hand insights from relevant conferences on these topics. The second will discuss the economic and regulatory aspects of AI. We will examine government regulations, policies, and standards from around the world and key documents such as the AI Index, reports from global firms, and other relevant documents. The third will examine the technical aspects of AI governance, including safety, robustness, and security. We will examine outputs and insights from leading institutions and thinkers in the field, and review existing best practices in the field. Each section will provide a brief overview of existing materials to inform participants of ongoing discourses in these fields, and offer questions for discussion. The following discussion will be themed around the relevance and application of these global discourses to Asia, with an emphasis on East and Southeast Asia.

Class Description (by Sunyoung Yang)

Applicability of Existing Global Governance Models for the Formation of Global AI Governance

Recent innovations in Artificial Intelligence (AI) have been accelerated by computing technologies and big data. AI can be defined as “activity devoted to making machines intelligent, and intelligence is that quality that enables an entity to function appropriately and with foresight in its environment” (Nilsson 2010). AI is both creating new frontiers for social experience, public infrastructure, and industrial innovation, but it also threatens to accelerate automation, social alienation, and global inequities. The impact of AI has worldwide implications, which calls for globally collaborative efforts to develop a joint governance of AI, as well as the need to allow diverse groups to contribute to the process. New regulations for self-driving cars and aircrafts, imposition of taxes for robot labor, and banning of autonomous weapons are examples where AI governance is needed. This lecture intends to articulate the components required to develop a global governance of AI based on the model of multi-stakeholder processes (Hemmati 2002). This lecture will determine what factors will be pertinent to AI governance from existing examples of global governance, including Internet Governance, the Intergovernmental Panel on Climate Change, more broadly the UN Environment, and global governance of nuclear technology. This lecture will then synthesize the existing examples of global governance and analyze the applicability of those examples for AI governance, examining Internet governance and AI governance within the broader context of digital governance.

Thematic Pillars (by Partnership of AI)

Safety-critical AI

Fair, Transparent and Accountable AI

Collaboration between People and AI

AI, Labor and Economy

Social and Societal Influences of AI

AI and Social Good

Lecture Candidates: Jaeseung Jeong, Sunyoung Yang, Danit Gal,...

References

AAAI, Spring Symposium Series, Palo Alot, 2018.3.26-28.

Acceptable Intelligence with Responsibility (AIR)

Philippe Aghion, Artificial Intelligence and Economic Growth, 2017.10.