Beyond EVE

AI and Content Moderation

Sun, 22 Feb 2026 13:12:37 +0100

Public pressure on platform companies to more soundly monitor the content on their sites is constantly increasing. To address this, platforms are turning to algorithmic content moderation systems. These systems prioritize content that promises to increase engagement and block content that is deemed illegal or is infringing the platform's own policies and guidelines. But content moderation is a ‘wicked problem’ that raises many questions all of which eschew simple answers. Where is the line between hate speech and freedom of expression – and how to automate and deploy this on a global scale? Are platforms overblocking legitimate content, or are they rather failing to limit illegal speech on their sites?

Within the framework of a ten-week virtual research sprint hosted by the HIIG, thirteen international researchers from various disciplines came together to tackle the challenges posed by automation in content moderation. Their work resulted in policy briefings focused on algorithmic audits and on increasing the transparency and accountability of automated content moderation systems. We warmly invite you to learn more about their findings and attend their output presentations followed by a panel discussion.

Agenda

Opening remarks on the project and the research sprint by research director Wolfgang Schulz and research lead Alexander Pirang

Presentations of the research outputs by the sprint fellows:

David Morar, guest researcher at George Washington University, Elliott School of International Affairs, USA
Aline Iramina, PhD candidate at the University of Glasgow, Great Britain
Sunimal Mendis, lecturer at the University of Tilburg, Netherlands

Followed by a panel discussion moderated by Jennifer Boone with:

Angelica Fernandez, fellow of the research sprint and PhD candidate at the University of Luxembourg
Philipp Otto, founder and director of the iRights.lab
Matthias Kettemann, associated researcher at the HIIG and scientific lead of the research project ”Regulatory Structures and the Emergence of Rules in Online Spaces” at the Leibniz-Institut für Medienforschung I Hans-Bredow Institut

Research for the Networked Society

Research for the Networked Society — Fri, 04 Dec 2020 14:44:35 +0100

The Weizenbaum Institute investigates the current changes in all aspects of society occurring in response to digitalisation. Our goals are to develop a comprehensive understanding of these changes based on rigorous academic analysis and to offer informed strategies to address them at a political and economic level. The Institute’s core objective is to conduct outstanding, interdisciplinary, and problem-oriented basic research, which at the same time drives application-oriented projects and, moreover, stimulates the formulation of new research questions. To do justice to the interplay between technology and society, the principle of interdisciplinarity will be implemented not only selectively, but in all research areas and projects. For the first time, the Institute will unite all relevant disciplines in a single research program and develop a holistic perspective on the process of digitalisation in society. A central social challenge is to ensure democratic self-determination and participation under the conditions of increasing digitalisation and automation. Accordingly, the Institute’s overarching question is the following: How can the goals of individual and social self-determination be achieved in a world characterised by digitally mediated processes of transformation and demarcation, and which framework conditions and resources are necessary for their realization? Here, self-determination is understood as the individual and collective competency to recognise, use, and design the scope of action. It is a fundamental prerequisite for the democratic organisation of society and a competitive market economy.

New York University - AI Now Institute

New York University - AI Now Institute — Thu, 03 Dec 2020 14:29:27 +0100

Artificial Intelligence systems are being applied to many arenas of human life – across major sectors such as education, health care, criminal justice, housing, and employment – influencing significant decisions that impact individuals, populations, and national agendas. But the vast majority of AI systems and related technologies are being put in place with minimal oversight, few accountability mechanisms, and little research into their broader implications. Currently there are no agreed-upon methods to measure and assess the social implications of AI, even as these systems are being rapidly integrated into core social institutions. To ensure that AI systems are sensitive and responsive to the complex social domains in which they are applied, we will need to develop new ways to measure, audit, analyze, and improve them. The AI Now Institute produces interdisciplinary research on the social implications of artificial intelligence and acts as a hub for the emerging field focused on these issues. Currently, our research focuses on four key domains: rights and liberties, labor and automation, bias and inclusion, and safety and critical infrastructure. Founded in 2017 by Kate Crawford and Meredith Whittaker, AI Now is housed at New York University, where it fosters vibrant intellectual engagement and collaboration across the University and beyond.

Is the United States Tax System Favoring Excessive Automation?

Information Technology & Innovation foundation ITIF — Sat, 02 Jan 2021 12:15:46 +0100

As the next wave of information technology matures, many commentators worry about the job disruption that automation technology will bring. In a recent policy brief released by the MIT Task Force on the Work of the Future, MIT economist Daron Acemoglu and his co-authors argued that the United States currently taxes machinery and equipment too little compared to labor, thereby encouraging excessive automation that eliminates jobs without making the economy more productive. ITIF President Rob Atkinson has argued in response that automation doesn’t lead to joblessness and that increasing taxes on automation equipment, including artificial intelligence, would hurt U.S. competitiveness and reduce real wage growth.

ITIF hosted a debate in which Acemoglu and Atkinson laid out their views about the future of automation technology and the effects it may have on U.S. competitiveness and the economy.

Speakers

Bernie Becker

Tax Reporter, POLITICO, Moderator

Robert D. Atkinson

President, Information Technology and Innovation Foundation, Speaker

Daron Acemoglu

Institute Professor, MIT, Speaker

The Future of Work: Preparing for Disruption

The World Bank — Sat, 05 Dec 2020 21:49:17 +0100

Workers of the future will need new sets of skills to compete. Recent advances in technology are changing how we live, communicate and do business, disrupting traditional industries and redefining the employee-employer relationship. Thousands of routine and low-skill jobs will be eliminated by automation, A.I. and digital hyper-connectivity. However, these same advances present new opportunities, like: - New job creation - Increased productivity - Improved delivery of public services This course emphasizes the urgency of developing human capital in meeting the challenges of the coming decades. Developing countries will need to take rapid action to invest in their people as innovation continues to accelerate. Themes in the course include: artificial intelligence, the gig economy, world of work, the future of work, labor market, policy makers, new technologies, digital economy, jobs of the future, machine learning, and labor force.

Auditing for Bias in Resume Search Engines

Berkman Klein Center for Internet & Society — Sat, 05 Dec 2020 20:38:41 +0100

There is growing awareness and concern about the role of automation in hiring, and the potential for these tools to reinforce historic inequalities in the labor market. In this work, Wilson and his team perform an algorithm audit of the resume search engines offered by several of the largest online hiring platforms, to understand the relationship between a candidate's gender and their rank in search results. They audit these platform with respect to individual and group fairness, as well as indirect and direct discrimination.

Christo Wilson's homepage here

Investigating the Impact of Gender on Rank in Resume by Le Chen, Ruijun Ma, Anikó Hannák and Christo Wilson

Ethics and algorithmic processes for decision making and decision support

Sat, 12 Dec 2020 21:54:14 +0100

Far from being a thing of the future, automated decision-making informed by algorithms (ADM) is already a widespread phenomenon in our contemporary society. It is used in contexts as varied as advanced driver assistance systems, where cars are caused to brake in case of danger, and software packages that decide whether or not a person is eligible for a bank loan. Actions of government are also increasingly supported by ADM systems, whether in “predictive policing” or deciding whether a person should be released from prison. What is more, ADM is only just in its infancy: in just a few years’ time, every single person will be affected daily in one way or another by decisions reached using algorithmic processes. Automation is set to play a part in every area of politics and law.

Current ethical debates about the consequences of automation generally focus on the rights of individuals. However, algorithmic processes – the major component of automated systems – exhibit a collective dimension first and foremost. This can only be addressed partially at the level of individual rights. For this reason, existing ethical and legal criteria are not suitable (or, at least, are inadequate) when considering algorithms generally. They lead to a conceptual blurring with regard to issues such as privacy and discrimination, when information that could potentially be misused to discriminate illegitimately is declared private. Our aim in the present article is, first, to bring a measure of clarity to the debate so that such blurring can be avoided in the future. In addition to this, we discuss ethical criteria for technology which, in the form of universal abstract principles, are to be applied to all societal contexts.

AI NOW 2017 Symposium

New York University - AI Now Institute — Fri, 04 Dec 2020 21:20:51 +0100

The AI Now 2017 Symposium was designed to address the biggest challenges we face as AI moves further into our everyday lives. This was the second annual Symposium hosted by the AI Now, with generous support from the AI Ethics and Governance Fund, John D. and Catherine T. MacArthur Foundation, and MIT Media Lab. The 2017 Symposium brought together over 100 leading experts from industry, academia, civil society, and government to share ideas for technical design, research, and policy directions. Discussions this year focused on the application of AI across four key themes: Rights and Liberties, Labor and Automation, Bias and Inclusion, and Ethics and Governance. These experts spent a day in closed-door talks and discussions, then joined an evening program that was free and open to the public. You can watch videos of the talks and panel discussions from both events below. The AI Now 2017 Report provides of summary of the Symposium’s four focus areas with close attention to developments that have occurred in the last 12 months. This 2017 Report also incorporates key insights and high-level recommendations that emerged from discussions at the Symposium.