CS 4732: Ethical and Social Issues in AI (Spring, 2017)

The course will meet for 8 weeks 7:30-9:55 PM on Monday evenings, starting Feb. 27. There will be a one-hour public lecture (which can be attended by anyone), followed by a discussion. The discussion is limited to the students registered in the course. The course will be pass/fail, but in order to pass, you will have attend all classes and participate actively. There will be short readings associated with all the presentations. Here is the current schedule of presentations:

• 2/27 - Bart Selman, Cornell: The Future of AI: Benefits vs. Risks

  • Reading: This interesting New York Times article provides good background to some of the current capabilities of AI systems.

• 3/6 - Jon Kleinberg, Cornell: Inherent Trade-Offs in Algorithmic Fairness

  • Abstract: Recent discussion in the public sphere about algorithmic classification has involved tension between competing notions of what it means for such a classification to be fair to different groups. We consider several of the key fairness conditions that lie at the heart of these debates, and discuss recent research establishing that except in highly constrained special cases, there is no method that can satisfy all of these conditions simultaneously. These results suggest some of the ways in which key notions of fairness are incompatible with each other, and hence provide a framework for thinking about the trade-offs between them.

  • Readings:
    • M. Luca, J. Kleinberg, S. Mullainathan. M. Luca, J. Kleinberg, S. Mullainathan. Algorithms Need Managers, Too. Harvard Business Review, 94:1 (Jan/Feb 2016).
    • J. Kleinberg, J. Ludwig, S. Mullainathan. A Guide to Solving Social Problems with Machine Learning. Harvard Business Review, December 2016.
  
  
• 3/13 - Kilian Weinberger, Cornell: Interpretable Machine Learning: What are the limits and is it necessary?
  • Abstract: Recent years have seen a revival of deep neural networks in machine learning. Although this has led to impressive reduction in error rates in some prominent machine learning tasks, it also raises the concern about interpretability of machine learning algorithms---can we understand/explain what they are doing. In this talk I will describe the basics of deep learning algorithms and explain their basic building blocks. I will show that these are easy to understand. I will also try to shed some light onto what these networks learn on a higher level and hopefully be able to convince the audience that these are not “black box” algorithms, as they are often described ... maybe instead gray or dark gray at most.

  • Reading: Christian Szegedy et al., Intriguing properties of neural networks

• 3/20 - Dan Weld, U. Washington: Computational Ethics for AI
  • Abstract: Stephen Hawking, Bill Gates, and other luminaries warn that an “intelligence explosion” may lead to the extinction of humanity at the hands of rampant robots. At the same time, many pundits see a prosperous future in which self-driving cars reduce highway fatalities , while AI advisors improve medical care and minimize malpractice. Weld argues that the utopian outcome is more likely, but only if we address several key social and technical challenges .

• 3/27 - Moshe Vardi, Rice: Humans, Machines, and Work: The Future is Now
  • Abstract: Automation, driven by technological progress, has been increasing inexorably for the past several decades. Two schools of economic thinking have for many years been engaged in a debate about the potential effects of automation on jobs: will new technology spawn mass unemployment, as the robots take jobs away from humans? Or will the jobs robots take over create demand for new human jobs? I will present data that demonstrate that the concerns about automation are valid. In fact, technology has been hurting working Americans for the past 40 years. The discussion about humans, machines and work tends to be a discussion about some undetermined point in the far future. But it is time to face reality. The future is now.

• 4/3 - spring break

• 4/10 - no lecture

• 4/17 - Karen Levy, Cornell: Working With and Against AI: Lessons from Low-Wage Labor
  • Abstract: As intelligent systems are increasingly deployed into workplaces, human workers are increasingly in positions to work in coordination and integration with them. To be sure, AI poses threats in the form of the replacement of workers -- but in addition, it stands poised to alter the content and quality of the work that remains. In this talk, I consider how human workers confront computational incursions into their workspaces, how we work alongside them -- and importantly, how we push back against them.

• 4/24 - Ross Knepper, Cornell: Autonomy, Embodiment, and Anthropomorphism: the Ethics of Robotics
  • Abstract: A robot is an artificially intelligent machine that can sense, think, and act in the world. Its physical, embodied aspect sets a robot apart from other artificially intelligent systems, and it also profoundly affects the way that people interact with robots. Although a robot is an autonomous, engineered machine, its appearance and behavior can trigger anthropomorphic impulses in people who work with it. In many ways, robots occupy a niche that is somewhere between man and machine, which can lead people to form unhealthy emotional attitudes towards them. We can develop unidirectional emotional bonds with robots, and there are indications that robots occupy a distinct moral status from humans, leading us to treat them without the same dignity afforded to a human being. Are emotional relationships with robots inevitable? How will they influence human behavior, given that robots do not reciprocate as humans would? This talk will examine issues such as cruelty to robots, sex robots, and robots used for sales, guard or military duties.

• 5/1 - Joe Halpern, Cornell: Moral Responsibility, Blameworthiness, and Intention: In Search of Formal Definitions

  • Abstract: The need for judging moral responsibility arises both in ethics and in law. In an era of autonomous vehicles and, more generally, autonomous AI agents, the issue has now become relevant to AI as well. Although hundreds of books and thousands of papers have been written on moral responsibility, blameworthiness, and intention, there is surprisingly little work on defining these notions formally. But we will need formal definitions in order for AI agents to apply these notions. In this talk, I take some preliminary steps towards defining these notions.

  More course-related material:

  • An interesting youtube video Humans Need Not Apply on how robots will take over our jobs.
  • Recent work out of Princeton shows that machine learning programs learn our biases (and thus will exhibit the same biases that we do when they make decisions). Here is a reasonable newspaper article on the work. Here is the original article from Science and a fairly accessible perspective on it, also from Science.