Sessions | Absolutely Interdisciplinary

A world of natural and artificial agents in a shared environment

May 7, 2024 | 10:00 AM – 12:00 PM | In person

Speakers: Gillian Hadfield, Peter Railton

As AI development pushes towards building autonomous AI agents, we can imagine a future, perhaps not too far off, during which humans and AI agents are interacting regularly in open domains. Clearly, such agents need to be able to cooperate with humans, just as humans have evolved to be highly cooperative with one another. In this session, Peter Railton and Gillian Hadfield will discuss what it might mean to develop AI agents that possess moral or normative competence, and whether there is any difference between those two ideas about competence.

Session: Algorithms to support AI safety cases

May 7, 2024 | 1:00 PM – 2:00 PM | In person

Speakers: Ashton Anderson (moderator), Roger Grosse

With the rapid advancement of large language model capabilities, it becomes increasingly essential to demonstrate that a given AI system doesn’t pose a catastrophic risk. In this session, Roger Grosse will outline how AI risks can be categorized into “AI Safety Levels” and the logic of how one might build a safety case at a given level. This motivates the need for algorithmic advances that will help build safety cases (or determine if a model is unsafe). Grosse will overview some recent work on training data attribution (TDA), where one tries to determine which training examples are responsible for a model’s outputs. Better TDA methods should help build safety cases by measuring the consequences of each stage of training for the model’s behavioral proclivities.

designing human-machine coexistence

May 7, 2024 | 2:30 PM – 4:30 PM | In person

Speakers: Huili Chen, I. Glenn Cohen, Anna Su (moderator)

Technological advances in artificial intelligence will not only potentially revolutionize human abilities but also facilitate the development of artificial entities capable of interacting with humans in both physical and virtual environments. How should we design and facilitate social and technical systems in an age of human and machine coexistence and still preserve the distinctiveness of humanity? In this session, I. Glenn Cohen examines developments in the field of medical AI and the law while Huili Chen considers the architectures and design choices that shape how humans interact with machines and robots.

Navigating the AI landscape: Insights from the AI Index Report and Global Public Opinion on AI Report

May 8, 2024 | 10:00 AM – 12:00 PM | In person

Speakers: Peter Loewen, Kelly Lyons (moderator), Ray Perrault

What will a future landscape shaped by AI and data look like? What challenges do we face today amidst the implementation of AI systems across society? This session will explore key insights from Stanford HAI’s latest AI Index Report and the Schwartz Reisman Institute’s forthcoming Global Public Opinion on AI Report, exploring the goals, methodologies, and applications of these reports, as well as key takeaways for researchers and policymakers. What findings have the greatest potential to help us build a beneficial future? Speakers will discuss globally emerging trends and strategies for navigating evolving challenges in analyzing the role of AI systems, their social impacts, and their public perception.

AI and the future of democracy

May 8, 2024 | 1:00 PM – 2:00 PM | In person

Speakers: Beth Coleman (moderator), Harper Reed

As the world approaches the biggest election year in recent history, the role of new technologies in shaping governance and politics is both omnipresent and ominous. This panel delves into urgent issues surrounding the potential sway of voters through AI and the looming threats of AI-generated videos, “rumour bombs,” and disinformation campaigns that have been enabled by the rise of powerful new AI tools. How can we protect established democratic and participatory systems from the misuse of generative AI? Key levers may include new technical solutions for improved facilitation and sensemaking of public engagement and democratic process. More controversially, other thinkers have proposed that “responsible” AI in this domain must move from consumer application to part of societal constitution and infrastructure. By unpacking the challenges and proposing solutions, the panel will contribute to ongoing discourse on shaping a future where technology supports, rather than undermines, democratic principles.

AI adoption in industry

May 8, 2024 | 2:30 PM – 4:00 PM | In person

Speakers: Avi Goldfarb (moderator), Kristina McElheran, Christy Prada

While recent headlines have breathlessly proclaimed that AI technologies are taking over the world of work, the realities of AI adoption in practice are much more nuanced and varied. As history demonstrates, the transformative effects and implications of new technologies often take considerable time—sometimes a generation or more—to be implemented at a system level. However, when these transformations take hold, the results can alter not only the structure of industries and economies but the surrounding culture as well. This session will explore what challenges exist in getting industries to adopt new technologies, and how those challenges can be overcome, emphasizing the need for collaborative research across disciplines and the development of new paradigms and workflows to unlock the potential of AI innovation.

Ian P. Sharp Lecture: Beth Simone Noveck, “From ballots to bots: AI’s transformative role in democratic societies”

May 8, 2024 | 5:00 PM – 8:30 PM | In person

Speaker: Beth Simone Noveck

The Ian P. Sharp Lectureship was established at the Faculty of Information Studies at the University of Toronto in 1989 through a gift from I.P. Sharp Associates Ltd. (A Reuters Company). It is intended to bring internationally renowned individuals to the campus to explore the transformative effects of information practice. The lectures, which are open to the profession and members of the public, are delivered every three to four years by a distinguished figure in information science and related fields.

In 2024, the lecture returns through a partnership between the Faculty of Information and the Schwartz Reisman Institute for Technology and Society. The 2024 The Ian P. Sharp Lecturer is Beth Simone Noveck, a professor at Northeastern University where she directs the Burnes Center for Social Change and its partner project The GovLab. Noveck is faculty at the Institute for Experiential AI, School of Law, and in the College of Social Sciences and Humanities, the College of Arts, Design, and Media, the College of Engineering, and affiliated faculty at the Khoury College of Computer Sciences.

The lecture will be hosted as the closing session for Absolutely Interdisciplinary 2024, SRI’s annual academic conference. The lecture will be followed by a reception. Admission is free, but registration is required. Everyone is welcome.

2024 Graduate Workshop

Interdisciplinary Dialogues on AI

How can technology, ethics, and collaboration drive a better future? What steps are required to ensure that the technical systems we design reflect our values as a society?

The Schwartz Reisman Institute for Technology and Society’s 2023–24 cohort of graduate fellows present “Interdisciplinary Dialogues on AI,” a special one-day workshop that will explore innovative solutions for tough problems at the intersection of technology and society. Across four thought-provoking sessions, a diverse range of scholars and practitioners will discuss exciting new research directions and address some of the most pivotal issues shaping our world today.

In the morning, two workshop sessions will examine the transformative potential of AI in healthcare delivery and the ethical dimensions of AI safety, navigating the complexities, challenges, and interdisciplinary opportunities that lie ahead for aligning AI systems with human values, real-world applications, and mitigating potential harms.

In the afternoon, we’ll engage in discussions surrounding online safety and content moderation, as our first session explores interventions to combat misinformation, polarization, and online violence in today's digital landscape. The following session will address the urgent need for sustainable urban development amidst climate change, and envision resilient cities through the lenses of technology and interdisciplinary collaboration.

Don't miss this unique opportunity to learn from an impressive lineup of leading thinkers who are shaping the future of research on technology’s social impacts and helping to define how powerful new tools like AI can benefit the world.

While in-person attendance at “Interdisciplinary Dialogues on AI” is by invitation only, registration for online participation is free and open to the public. We hope you’ll join us!

SRI’s graduate workshop is coordinated by: Michael Colacci, Felix Menze, Jo-Ann Osei-Twum (Innovating care); Michael Beauvais, Jamie Duncan, Kelly McConvey, Silviu Pitis, Michael Zhang (Harming virtuously); Alice Huang, Blake Lee-Whiting, Ramaravind Kommiya Mothilal, Emily Schwartzman (Online safety); Taneea S. Agrawaal, Hiu-Fung Chung, Wenxi Liao, Han Qiao (Changing climate). Workshop coordination is supervised by Postdoctoral Fellows Atrisha Sarkar and Hayane Dahmen.

Innovating care: An interdisciplinary dialogue on AI in healthcare

May 6, 2024 | 9:10 AM – 10:40 AM | In-person and online

Speakers: Dr. Mamatha Bhat, Daniel Buchman, Muhammad Mamdani, Luke Stark (moderator)

In this session, speakers will present real-world examples to discuss the assumptions underpinning the use of artificial intelligence (AI) systems in healthcare, raise key considerations from their disciplinary perspectives for addressing issues in healthcare delivery, and explore current challenges and tensions within the field of AI and health. Attendees will gain a better understanding of the interdisciplinary challenges, tensions, and potential opportunities for the practical use of AI technologies in healthcare.

Harming virtuously: Value alignment for harmful AI

May 6, 2024 | 11:00 AM – 12:30 PM | In-person and online

Speakers: Branka Marijan, Leah West, Nisarg Shah

The interdisciplinary fields of AI safety and AI ethics emphasize the need to create autonomous systems that align with human values. These discussions often assume that AI systems should “do no harm” to humans. But lethal AI systems used for autonomous firearms and drones are already being used to harm people. In this session, we will explore how researchers and practitioners interested in the problem of AI alignment can address the reality of purposely harmful AI systems.

What is the future of online safety and content moderation?

May 6, 2024 | 1:30 PM – 3:00 PM | In-person and online

Speakers: Ishtiaque Ahmed, Bree McEwan, Regina Rini

Online discussions have, in recent years, become one of the main ways in which individuals access information and exchange opinions. Traditional barriers to communication have largely disappeared, and it is increasingly difficult to verify the sources and reliability of online information. In light of problems related to polarization, misinformation, online violence, and deepfakes, various interventions have been proposed. This session will facilitate a discussion about important questions related to these interventions.

Changing climate: The potential and limits of technology for sustainable cities

May 6, 2024 | 3:20 PM – 4:50 PM | In-person and online

Speakers: Maaz Gardezi, Vanessa Gray, M. Reza Najafi

The looming threat of climate change has heightened concerns regarding how complex urban systems—shaped by local, national, and transnational dynamics—address environmental challenges. Environmental data plays a pivotal role in visualizing the “climate problems,” yet its production, collection, and use are essentially contested. To understand the potential and politics of environmental data, adopting an interdisciplinary and cross-sectoral approach is crucial. Centered on the Greater Toronto Area, this session invites scholars interested in the use of data and AI for tackling urgent global problems to reimagine the intersection of climate change, data-driven technology, and socio-ecological justice, envisioning the potential for building resilient cities.