רגולציות, מסגרות, עקרונות והמלצות לאלגוריתמיים אתיים AI ethics and regulation

רגולציות, מסגרות, עקרונות והמלצות לאלגוריתמיים אתיים

AI ethics and regulation

Denisa Reshef Kera| STS
278614-01

Course description and learning goals

This module acts as an “observatory” following the current developments in emerging technologies and regulations related to AIs, GAIs, and LLMs including their convergences with blockchain technologies. We will follow various emerging initiative and policy documents that propose regulations and recommendations on ethical use of AIs and compare them with existing initiatives. The module maps the various attempts by private companies, research institutions, and the public sector (government and intergovernmental organizations) to define principles and guidelines for artificial intelligence (AI) and machine ethics. We will discuss how and at what stage of algorithm development the recommendations introduce external control and how they respond to the challenges of fully automated AI development without human control. We will trace how the various documents define responsible AI development and risk, how they create incentives, engage stakeholders, and resolve issues of communication and inclusion, how they operationalize the principles they identify, and how they monitor impact. The goal of the course is to introduce students to key policy texts (white papers, recommendations, standards), support critical discussions about their implementation, and assess their impact on data- and algorithm-driven society.

Learning objectives 

Knowledge:

Familiarize with main AI ethics documents and different approaches to ethics and regulation of algorithmic services. Understand the implications of biases, discriminations, and risks in algorithmic services. Learn to analyze and compare various ethical frameworks and recommendations for AI and algorithms.

Skills:

Develop the ability to read, analyze, and critically evaluate the strategies of different regulators and institutions. Gain proficiency in formulating policy recommendations based on the study of various frameworks. Acquire skills to define and analyze the strengths and weaknesses of different regulatory approaches and apply this knowledge to real-world cases.

Values:

Cultivate a deep understanding of the ethical, social, and policy implications of AI and algorithms. Promote the importance of socially responsible and ethical AI development.

Encourage critical thinking about the role and impact of AI in society and the importance of inclusive and effective regulation.

Lessons plan (Including active learning: 

Final grade

Active learning in this module includes:

Interactive Lectures and Discussions:

Host guest lectures from experts in AI regulation and ethics.

Facilitate class debates on controversial AI policies and their societal impacts.

In class Group Projects and Collaborative Work:

Assign team projects to analyze and compare different AI ethics documents and regulatory approaches. Encourage group discussions to formulate new policy recommendations or briefs.

Case-Based Learning:

Use real-world case studies to explore the implications of various AI policies and ethical frameworks. Engage students in role-playing exercises, simulating negotiations or discussions between different stakeholders in AI development and regulation.

In class Workshops and Simulations:

Conduct policy drafting workshops where students create their own AI ethics guidelines or regulatory frameworks. Implement simulations of AI regulation implementation and impact assessment.

Research and Analysis Assignments:

Assign research papers focusing on analyzing specific AI ethics documents or regulatory approaches. Encourage comparative analysis of different regulatory models and their effectiveness.

Presentations and Peer Review:

Have students present their policy briefs or recommendations and receive peer and instructor feedback. Organize student-led seminars on selected topics related to AI ethics and regulation.

Through this comprehensive approach, the module aims to equip students with a deep understanding of the current landscape of AI ethics and regulations, fostering critical thinking and practical skills in shaping and analyzing policy in the rapidly evolving field.

Course requirements

Participation in class discussions, in-class assignments, and peer assessment, including opening and leading a discussion topic (week): 50%

Students will be expected to read the papers before class, participate in discussions, and engage in in-class assignments (including peer assessment and feedback). They will individually select a week to open and lead class discussion by preparing a 20-minute presentation summarizing the main questions and positions of the papers assigned for that week. They will be encouraged to find a creative way to present the material and engage the class with 3-5 discussion points.

Final paper (reflection of the process and results): 50%

Assessment criteria include the ability to meet the formal criteria of a policy memo (explained in class and separate document), define and address the appropriate audience, provide empirical evidence (based on research of papers and data), provide a comprehensive, balanced, and focused argument, clearly defined scope, and recommendations.

Prerequisites

no

Bibliography: Up-to-date reading, viewing, and listening content items

A European approach to artificial intelligence | Shaping Europe’s digital future. (n.d.). Retrieved September 3, 2022, from https://digital-strategy.ec.europa.eu/en/policies/european-approach-artificial-intelligence 

Act. (2021, February 10). The Artificial Intelligence Act. https://artificialintelligenceact.eu/the-act/ 

Algorithmic Accountability for the Public Sector Report. (n.d.). Retrieved September 4, 2022, from https://ainowinstitute.org/pages/algorithmic-accountability-for-the-public-sector-report.html 

AI at Google: Our principles. (2018, June 7). Google. https://blog.google/technology/ai/ai-principles/

Beijing: Harmonious Artificial Intelligence Principles. (n.d.). Retrieved September 4, 2022, from https://bii.ia.ac.cn/hai/ 

Blackman, R. (2020, October 15). A Practical Guide to Building Ethical AI. Harvard Business Review. https://hbr.org/2020/10/a-practical-guide-to-building-ethical-ai 

Center for Democracy and Technology: Digital Decisions - https://cdt.org/wp-content/uploads/2018/09/Digital-Decisions-Library-Printer-Friendly-as-of-20180927.pdf 

Corrêa, N., Galvão, C., Santos, J., Del Pino Carvalho, C., Pinto, E., Barbosa, C., Massmann, D., Mambrini, R., Galvão, L., & Terem, E. (2022). Worldwide AI Ethics: A review of 200 guidelines and recommendations for AI governance. https://doi.org/10.48550/arXiv.2206.11922

Demystifying Explainable Artificial Intelligence: Benefits, Use Cases, and Models. (n.d.). Birlasoft. Retrieved September 4, 2022, from https://www.birlasoft.com/articles/demystifying-explainable-artificial-intelligence 

 

Fjeld, J., Achten, N., Hilligoss, H., Nagy, A., & Srikumar, M. (2020). Principled Artificial Intelligence: Mapping Consensus in Ethical and Rights-Based Approaches to Principles for AI (SSRN Scholarly Paper No. 3518482). https://doi.org/10.2139/ssrn.3518482 

More resources https://cyber.harvard.edu/publication/2020/principled-ai 

Hauer, T. (2022). Importance and limitations of AI ethics in contemporary society. Humanities and Social Sciences Communications, 9(1), 1–8. https://doi.org/10.1057/s41599-022-01300-7
IEEE Ethics In Action in Autonomous and Intelligent Systems | IEEE SA. (n.d.). Retrieved September 3, 2022, from https://ethicsinaction.ieee.org/ Document https://ethicsinaction.ieee.org/wp-content/uploads/ead1e.pdf 

Jobin, A., Ienca, M., & Vayena, E. (2019). The global landscape of AI ethics guidelines. Nature Machine Intelligence, 1(9), 389–399. https://doi.org/10.1038/s42256-019-0088-2  Access https://arxiv.org/ftp/arxiv/papers/1906/1906.11668.pdf 

Kazim, E., & Koshiyama, A. S. (2021). A high-level overview of AI ethics. Patterns, 2(9), 100314. https://doi.org/10.1016/j.patter.2021.100314 

Latonero, M. (2018, October 10). Governing Artificial Intelligence. Data & Society; Data & Society Research Institute. https://datasociety.net/library/governing-artificial-intelligence/ 

Learning, T. I. for E. A. & M. (n.d.). The Institute for Ethical AI & Machine Learning. Retrieved September 4, 2022, from https://ethical.institute 

Montreal University: The Declaration—Montreal Responsible AI. (n.d.). Respaideclaration. Retrieved September 4, 2022, from https://www.montrealdeclaration-responsibleai.com/the-declaration 

OECD Framework for Classifying AI Systems to assess policy challenges and ensure international standards in AI. (n.d.). Retrieved September 3, 2022, from https://oecd.ai/en/wonk/classification

OECD. (2019). Scoping the OECD AI principles: Deliberations of the Expert Group on Artificial Intelligence at the OECD (AIGO). OECD. https://doi.org/10.1787/d62f618a-en 

OECD Artificial Intelligence (AI) Principles. (n.d.). Retrieved September 3, 2022, from https://oecd.ai/en/ai-principles 

OpenAI Charter. (n.d.). OpenAI. Retrieved September 4, 2022, from https://openai.com/charter/ 

Principles for Accountable Algorithms and a Social Impact Statement for Algorithms: FAT ML. (n.d.). Retrieved September 3, 2022, from https://www.fatml.org/resources/principles-for-accountable-algorithms 

Proposal for a Regulation laying down harmonised rules on artificial intelligence | Shaping Europe’s digital future. (n.d.). Retrieved September 3, 2022, from https://digital-strategy.ec.europa.eu/en/library/proposal-regulation-laying-down-harmonised-rules-artificial-intelligence 

 

Salo-Pöntinen, H. (2021). AI Ethics—Critical Reflections on Embedding Ethical Frameworks in AI Technology. Culture and Computing. Design Thinking and Cultural Computing: 9th International Conference, C&C 2021, Held as Part of the 23rd HCI International Conference, HCII 2021, Virtual Event, July 24–29, 2021, Proceedings, Part II, 311–329. https://doi.org/10.1007/978-3-030-77431-8_20

State of AI Ethics. (2020, December 8). Montreal AI Ethics Institute. https://montrealethics.ai/state/
https://montrealethics.ai/volume6/  

Stanford report: Government by Algorithm: Artificial Intelligence in Federal Administrative Agencies
https://www-cdn.law.stanford.edu/wp-content/uploads/2020/02/ACUS-AI-Report.pdf